Polylactic Acid Nanofiber Scaffold Decorated with ... - ACS Publications

May 24, 2017 - Polylactic Acid Nanofiber Scaffold Decorated with Chitosan Islandlike. Topography for Bone Tissue Engineering. Ting Xu,. †,§. Hongya...
0 downloads 0 Views 3MB Size
Subscriber access provided by Binghamton University | Libraries

Article

Polylactic Acid Nanofibers Scaffold Decorated with Chitosan Island-Like Topography for Bone Tissue Engineering Ting Xu, Hongyang Yang, Dongzhi Yang, and Zhong-Zhen Yu ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 24 May 2017 Downloaded from http://pubs.acs.org on May 24, 2017

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

ACS Applied Materials & Interfaces 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 36

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

Polylactic Acid Nanofibers Scaffold Decorated with Chitosan Island-like Topography for Bone Tissue Engineering Ting Xu, a a



Hongyang Yang, a



Dongzhi Yang,*a

ZhongZhen Yu*a,b

State Key Laboratory of Organic-Inorganic Composites, College of Materials Science and

Engineering, Beijing University of Chemical Technology, Beijing 100029, China b

Beijing Advanced Innovation Center for Soft Matter Science and Engineering, Beijing

University of Chemical Technology, Beijing 100029, China ⊥

These authors contributed equally to this work.

E-mail: [email protected] (D.-Z. Yang); [email protected] (Z.-Z. Yu)

ABSTRACT: In this work, a bicomponent scaffold with a core-shell and island-like structure that combines the respective advantages of polylactic acid (PLA) and chitosan (CS) was prepared via electrospinning accompanied by an automatic phase separation and crystallization. The objective of this research was to design nanosized topography, and the high bioactivity of chitosan onto PLA electrospun fibers surface to improve the cell biocompatibility of PLA fibrous membrane. The morphology, inner structure, surface composition, crystallinity, and thermodynamic analysis of nanofibers with various PLA/CS ratios were respectively analyzed, and the turning mechanism of a core-shell or island-like topography structure was also speculated. The mineralization of hydroxyapatite and culture results of preosteoblast (MC3T3-E1) cells on the modified scaffolds indicate that the outer CS component and rough nanoscale topography on the surface of the nanofibers balanced the hydrophilicity and hydrophobicity of the fibers, enhanced their mineralization ability, and made them more beneficial for the attachment and growth of cells. Moreover, chitosan and “island-like”

1 Environment ACS Paragon Plus

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

protrusions on the fiber surface increased the ALP activity of the MC3T3-E1 cells seed on the fibrous membrane, and provided more appropriate interface for cell adhesion and proliferation. These results illustrate that this kind of PLA/CS membrane has the potential in tissue engineering. More importantly, our study provides a new approach to design PLA scaffolds with combined topographic and bioactive modification effects at the interface between cells and materials for biomedicine. KEYWORDS: electrospun, polylactic acid, chitosan, core-shell, island, tissue engineering

1. INTRODUCTION In recent years, electrospinning has become a hottest technology for the preparation of tissue engineering scaffolds owing to its facility to produce nanofibrous scaffolds with highly specific surface areas and porosities, and with structural features that have a bionical morphology of the ECM.1-2 Compared to other scaffolds, electrospun nanofibers with isotropic aligned fibers have shown a superior ability to guide cell adhesion and migration, affect cell spreading and differentiation.3 Synthetic polyesters is a widely used raw materials for electrospun tissue engineering scaffolds. For example, PLA,4-5 PCL6-7 and PLGA 8-9 have already been approved by the FDA to be widely applied in the biomedical field because of their excellent degradability, mechanical properties, biocompatibility and good design control property. However, the cell affinity toward these polyester scaffolds is not satisfactory due to their higher hydrophobicity, and few bioactive sites, as compared with natural ECMs. Therefore, based on the advantages of electrospun fibrous membrane simulating the structure of natural ECMs, it has been a challenge for polyester to achieve suitable cell adhesion and surface growth.

2 Environment ACS Paragon Plus

Page 2 of 36

Page 3 of 36

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

Commonly, natural polymers10-15 with good bioactivity are preferred as modification components because they can provide cell recognition sites, and render innate biological information guidance to cells. For example, electrospun chitosan scaffolds with good bioactivity have been broadly investigated in the tissue engineering field.16-18 However, it is generally known that their poor mechanical properties and rapid degradation rates limit their application.19-20 In recent years, the blending nanofibers combining the advantages of PLA and CS has attracted the attention of researchers.21 For example, Nguyen 22 and Wang et al 23 prepared PLA/CS core-shell nanofibers; while Shalumon, K. T. et al

24

reported that aligned PLA/CS blended nanofibers

guided the aligned growth of human dermal cells. Au, H. T. et al

25

focused on PLA/CS

composite nanofibers with outstanding antibacterial activity for wound-healing applications; and Yu-Ru, Lee et al

26

successfully obtained a guided bone regeneration membrane (GBR) made of

hydroxyapatite mineralized on a CS-coated electrospun PLA nanoweb. Various approaches have been demonstrated for the surface treatment of scaffolds, such as coating, layer-by-layer technologies,27 grafting via ray irradiation,28 and electrostatic interaction et al.29 For electrospun fibrous membrane, coaxial electrospinning is a common method for the preparation of multi-component nanofibers with a desirable functional surface.30 It is worth mentioning that a good match is desired between the rheological properties of the inner and outer components for continuous and uniform core-shell nanofibers. Besides coaxial electrospun technologies, single nozzle electrospinning with a homogeneous or emulsion solution is another facile method to obtain core-shell structures. However, this method is only favorable for binary incompatible polymers in the same solvent system. It is well known that besides scaffold components, topography structure is also important factor that affects the adhesion, growth, and cell differentiation. In addition, it has also been a factor

3 Environment ACS Paragon Plus

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

ascribed to improve protein adsorption.31 Theoretically, the simultaneous use of (i) outer CS bioactive components and (ii) a nanosized rough topography on the PLA nanofibers was more effective at fine-tuning the scaffold’s biocompatibility, than using core-shell nanofibers with a smooth surface. Because of the automatic phase separation mechanism between two incompatible polymers during the electrospinning process, it is possible to switch the structure of the fibers from core-shell to “island-like” with nanoscale protrusions. In our previous work, core-shell PEO/CS bicomponent nanofibers were first reported by a homogenous solution, and the formation mechanism was inferred the binodal phase separation of ternary system in the fiber splitting process, accompanied by solvent evaporation.32 Similarly, for PLA/CS binary fibers, the size and shape of the chitosan island could be tuned by the solvent evaporation speed, and PLA crystallization process. A convenient and simple control of the electrospun temperature served this purpose. Chitosan island-structured PLA nanofibers have the following essential advantages for improving the adaptation of cells to the scaffold surface: (1) a controllable balance of the surface hydrophilicity and hydrophobicity; (2) suitable roughness at the surface for cell attachment; (3) cell recognition sites provided by the CS component, which renders innate the biological information guidance to the cells. Furthermore, it is a facile and effective method that could be extended to other binary polymer scaffolds to improve the surface behavior of bulk materials. Our results demonstrate that this structure significantly enhanced the mineralization of hydroxyapatite and the attachment and growth of preosteoblasts (MC3T3-E1) on the scaffold’s surface, as we had expected. In addition, to demonstrate the formation mechanism of core-shell and island-like structures on the surface of the fibers, the morphology, inner structure, surface composition, crystallinity, and thermodynamics of the nanofibers were studied in detail. The bone-bonding ability of the scaffolds was evaluated by investigating hydroxyapatite

4 Environment ACS Paragon Plus

Page 4 of 36

Page 5 of 36

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

mineralization in SBF. In addition, the behavior of MC3T3-E1 preosteoblast cells grown on the scaffolds was also investigated to explore their feasibility used in tissue engineering.

2. EXPERIMENTAL SECTION Materials. Polylactic acid (PLA) with Mw of 150000 g/mol was obtained from Shanghai Biodegradable Materials Technologies Co. Ltd. (Shanghai, China). Chitosan (DA 90%) was supplied by Zhejiang Golden Shell Co. (Zhejiang, China). Trifluoroacetic acid (TFA) and all other chemicals for the preparation of the SBF were supplied by Beijing Chemical Reagent Company (Beijing, China). All the above mentioned reagents were used without further purification. The mouse preosteoblasts (MC3T3-E1) from bone-calvaria were purchased from the Peking University Health Science Center (Beijing, China). α-Minimum essential medium (α-MEM), ethylenediamine tetraacetic acid (EDTA), fluorescein isothiocyanate (FITC), and other essential reagents for the cells culture tests were supplied by Sigma-Aldrich Trading Co., Ltd. (Shanghai, China) Fabrication of Nanofibers by Electrospinning. The core-shell and island-like structure of the nanofibers was regulated by the phase separation of PLA and CS in PLA/CS homogenous solution via electrospinning. Our preparation process is shown in (Scheme 1). The detailed process was as follows: 22 wt.% of CS and PLA were dissolved in TFA to form a homogeneous blending solution, using different CS/PLA weight ratios varying from 90:10 to 10:90. Pure CS and PLA solutions were used for comparison. The above solutions in a 10 mL plastic syringe was fitted with a metal nozzle (with an inner diameter of 0.60 mm), and the collecting distance was set at 20 cm. A 20 kV voltage, and a spinning rate of 1.60 mL/h were applied. A perfect Taylor cone formed on the needle tip, and well-electrospun PLA/CS nanofibers were collected onto the aluminum foil. Unless

5 Environment ACS Paragon Plus

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

Page 6 of 36

otherwise noted, the electrospinning temperature was controlled at 25°C, 30% relative humidity, collective distance 15 cm and solution flow rate 1.6 mL/h, respectively. For investigation the effect of the temperature on the fiber surface morphology, and the phase separation of PLA and CS, some electrospun experiments were carried out at 35, 45, 50, 55, and 60°C for comparison. Finally, the nanofibers were dried at 40°C in a vacuum oven for 24 h. And there is also another extraction process for removing the solvent and residual crosslinking agent glutaraldehyde before cell culturing on the fibrous membrane, the fibrous membrane was extracted in distilled water for 48h and the medium was refreshed every 4-6h until the pH closed neutral.

Chitosan:

core-shell electrospinning O

PLA:

O n CH3

Phase separation process by solvent evaporation island

Sheme 1. The preparation process of CS islanded structured scaffolds

Fluorescence Labeling of CS. In order to prove that CS mainly existed in the shell of the nanofibers, the nanowebs (1 cm × 1 cm) were immersed in a 0.1 wt.% methanol solution of fluorescein dye FITC for 8 h in the dark (the volume ratio of methanol and deionized water was 10:4, pH 9). Then the membranes were rinsed with ethanol and distilled water to remove the unreacted FITC, dried, and analyzed by laser scanning confocal microscopy.33-34

6 Environment ACS Paragon Plus

Page 7 of 36

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

Mineralization of the Nanowebs with 10 × SBF. Mineralization of the PLA/CS nanowebs was carried in 10 × SBF after cross-linking with glutaraldehyde vapor (50 wt.% aqueous solution) for 3 h at 25°C. A saturated solution of 10 × SBF were prepared with 5.49 g NaCl, 0.36 g NaHCO3, 0.29 g CaCl2, 0.36 g Na2HPO4·12H2O and 100 mL deionized water. The PLA/CS nanowebs (3 cm × 3 cm) were immersed in above 10 × SBF solution for 3-24 h at 37°C and mild stirring. The SBF solution was refreshed at every intervals of 3 h. When taken out from the 10 × SBF solution, the samples were treated gently with deionized water, and then vacuum drying carried out at 40°C for 24 h. 35-36 Characterization. The surface morphology of the prepared nanofibers was characterized using a SEM (S-4700, Hitachi Company, Japan). The inner structure of the nanofibers was observed by a TEM (H-800, Hitachi Company, Japan). XPS (Escalab 250 Thermo Fisher Scientific Corporation, USA) was chosen to analyze the composition of the outer layer of the nanofibers. The crystalline behavior and thermodynamic behavior of the fibers were respectively investigated on a wide angle XRD analyzer (D8 ADVANCE, Bruker AXS Gmbh, Germany), and DSC (Q100 TA Instruments, USA), under 10/min rate from 30 to 100 °C. The elemental and content analysis of the mineral hydroxyapatite crystals deposited on the PLA/CS nanofibers was performed by an EDS attached to the SEM system. The surface hydrophilicity of the PLA/CS nanofibrous membranes was evaluated on a contact angle goniometer (DSA-30, Kruss, Germany), five sites on the same mambrane were detected. The fluorescence intensity of the FITC labeled nanofibers, and the growth and adhesion of the cells were observated by a LSCM (Leica SP5, Leica Microsystems Wetzlar GmbH, Germany). An exciton of the fluorophore with a 488 nm Ar-Kr laser was used to characterize the FITC labeled nanofibers.

7 Environment ACS Paragon Plus

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

Cells Culture. The mouse preosteoblasts MC3T3-E1 were cultured in α-MEM medium with 10% FBS, 1% penicillin/streptomycin, 3 mM β-glycerophosphate, and 10 mg/mL ascorbic acid at 37°C and in 5% CO2. The medium was refreshed every 24h. The cells were trypsinized by 0.25% trypsin with 1 mM EDTA when they reached 80% confluence. Cells Proliferation. The samples were sterilized in highly compressed steam for 20 min. After that, the MC3T3-E1 cells were seeded on the membrane with a density of 3×103 cells per well of 96-well plate. At 1, 2, 3, 7 and 14 days after culturing, the cell viability was detected with CCK-8 at 450 nm on an ELISA reader (Multiscan MK3, Labsystem Co. Finland). For comparison, α-MEM medium with 10% FBS and same medium with 1% phenol were chosen as the negative control and positive control, respectively. The cell number adhered on the nanofiber membrane was calculated according to the functional relation between the absorbance and cell numbers.3 Cells Spreading Morphology Observation by SEM and LSCM. To observe the attachment ability of the cells and their shapes on the different scaffolds, all scaffolds were separately fixed on glass coverslips, sterilized in a 70% ethanol solution overnight, then rinsed with a sterile phosphate buffer saline (PBS) solution three times, and finally placed into a 24-wells plate pre-treated with poly (HEMA) solution to avoid the attachment of the cells to the plates. Approximately, 3×103 trypsinized cells/mL were seeded on each sample and cultured for 1, 2, 3, 7, and 14 days. For their observation with SEM, 3.7% formaldehyde-PBS solution and ethanol were used to fix and dehydrate the cells, respectively. For LSCM observations, 0.1% Triton X-100 was used to improve the permeability of the cell’s membrane. Afterwards, cytoskeletal F-actin of cells were stained by a 50 µg/mL phalloidin labeled with TRITC for 40 min, and then rinsed with a PBS removing redundant dye. While the cell nuclei

8 Environment ACS Paragon Plus

Page 8 of 36

Page 9 of 36

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

were stained by Vectashield mounting medium with DAPI. The cell morphologies were visualized at 405 nm (DAPI) and 543 nm (TRITC) by an LSCM.37 ALP Activity Test and Osteocalcin Expression. ALP activity of the cells was tested after they had been seeded for 3, 7, 14, and 21 days according to the literature.32 Specifically, 400 µL of a p-nitrophenyl phosphate and disodium salt solution was dropt to each well and reacted for 1 h. Thereafter, 200 µL of a 2 M NaOH solution were transferred into each well using a pipettor. Finally, 100 µL of the reaction product were transferred to a blank 96-well microplate reader, then the absorbance was tested at 450 nm. For the Osteocalcin expression, cell lysate was detected by an Elisa kit at 450 nm after culturing the cells for 7 and 14 days. RESULTS AND DISCUSSION Structure and Morphology of Nanofibers. In this section, we focused on the effects of the PLA and CS component ratios, and the electrospinning temperature, on the inner and outer phase separation structure of the fibers. The surface morphology of nanofibers prepared at the same temperature, but with PLA/CS weight ratios varying from 0 to 100%, are displayed in (Fig. 1). This figure also shows the morphology of nanofibers prepared at different temperatures but with a fixed PLA/CS ratio of 70:30. The results indicate that pure PLA fibers with an average diameter of about 1µm showed a smooth and uniform surface morphology without any beads. PLA/CS fiber diameters decreased as the CS content increased, while the polydispersity increased. Chitosan fibers with 200 nm diameter were uneven and fragile.

9 Environment ACS Paragon Plus

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 10 Environment ACS Paragon Plus

Page 10 of 36

Page 11 of 36

1000

N Island dimension and distribution (nm)

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

800

periodic distance island size hight of island

600

400

200

0 35

50

60

Figure 1. SEM images of electrospun PLA/CS nanofibers. A-E) Fibers prepared using different PLA/CS weight ratios at room temperature. F) Histogram of the average diameter distribution of all electrospun nanofibers. G-I) Fibers prepared at different temperatures using a PLA/CS 70:30 weight ratio, the island distance increased with temperature increase. J) PLA/CS 80:20, 35 °C; K) PLA/CS 70:30, 35 °C, cross-linked, and etched with deionized water for 24 h, CS island remained on the fiber surface. L) PLA/CS 70:30, 35 °C, non-crosslinked, etched with deionized water for 24 h, fibers showed smooth surface after removing CS component. M) Schematic of the island-like structure. N) Island size and distribution. About island height and the vertical distance between islands, AFM analysis results in supporting information (Figure S-1) also confirmed SEM results. In (Fig. 2B-C and Fig. S-2 A-C), TEM micrographs display the clear interface boundary between the core and shell structure of the electrospun PLA/CS nanofibers, with varies CS ratios, owing to the phase separation of PLA and CS accompanied by the solvent evaporating. (Fig. 1G-I) show the morphology of electrospun PLA/CS 70:30 carried out at 35 °C, 50 °C and 60 °C. Compared to the smooth nanofibers obtained at 25 °C, these nanofibers became rough, and many small bumps with an “island-like” structure appeared on their outer surface. In order to confirm the major composite of the shell, the surface morphology of the fibers before and after

11 Environment ACS Paragon Plus

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

Page 12 of 36

cross-linking, and etched by deionized water, was also characterized by SEM. As shown in (Fig. 1K and L), for cross-linked sample, the “island” structure on the outer surface still existed, but for non-crosslinked sample, it disappeared. This phenomenon illustrates that the “island-like” CS structure appeared because CS is soluble in aqueous solutions, but PLA is not. In order to further illustrate the composition of the “sea” and “island” segments, electrospun PLA/CS 70:30 nanofibers at 35°C were collected on a copper net, and then etched with dichloromethane. The sharp contrast of the bright and dark regions in the TEM image (Fig. 2E) displays the morphology of a hollow interior but an external continuous “cellular” structure. Combined with the TEM micrograph analysis of the rinsed fibers with a core-shell and island-like structure, the phenomenon further demonstrated that the “sea” was made of PLA because dichloromethane dissolves PLA but not CS. Meanwhile, with the exception of PLA/CS 70:30, electrospun PLA/CS 80:20 nanofibers at 35°C also presented the similar “island-like” structure, as shown in (Fig. 1J). The above speculation could also be further proved by the XPS analyses. We speculated that the phase separation of PLA and CS occurred during the solvent evaporation step of the electrospinning process. One possible reason for the immigration of the CS phase to the surface of the fibers due to that CS with positively charged groups prefer to move out under the electrical field during electrospinning. Some eccentricity was found in (Fig. 2C), presumed to be caused by a larger number of positively charged CS units, causing some instability of the composite solution, and strong whipping during electrospinning. These results were easy understood by an increase of viscosity of the PLA/CS solution, accompanied by the evaporation of the solvent, allowing for PLA to block the aggregation of the CS molecules. In addition, the repulsive electrical force of intermolecular interactions also hindered CS from gathering. Therefore, the PLA “sea” and CS “island” structure formed.

12 Environment ACS Paragon Plus

Page 13 of 36

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

Furthermore, the size and distribution of CS-islands could be easily regular by the control of electrospinning temperature. As shown in (Fig. 1M and N), an increase of the electrospinning temperature accelerated the solvent evaporation rate, and the number of chitosan molecules that migrated to the surface of the fiber was reduced. This caused the periodic distance of the islands to increase, and the size of the islands to decrease. To optimize the fiber morphology and film strength, we selected the PLA/CS 70:30 composite ratio in the following experiment. Furthermore, the distribution and size of CS-islands could be easily regular by the control of electrospinning temperature (Figure 1M and 1N), for example, increasing the temperature from 35 ℃ to 60 ℃, the periodic distances of islands obviously increased from 400 to 750 nm, and the island size decreased from 200 to 140 nm. Besides, the topographical analysis of the fiber surface (AFM) was also used to characterize the construct of nanofiber (Figure S-1). The aligned PLA/CS island-like fibers were collected in the gap (3cm) between two pieces of magnetite at 35°C, and the island height and distance between the islands were recorded by AFM analysis software. The island height is about 50nm, the vertical distance between islands is about 700nm. This phenomenon will be explained in the follow section “Formation Mechanism of Core-shell and Island Structure”.

13 Environment ACS Paragon Plus

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

Page 14 of 36

Figure 2. TEM images of electrospun PLA /CS. A PLA; B PLA /CS 70:30; C PLA /CS 10:90; D CS; E PLA/CS70:30, 35 °C, etched by dichloromethane. Compared to neat PLA or CS, all composite PLA/CS fibers displays core-shell prepared under room temperature. Sharp contrast of the bright and dark regions in the TEM image displays the morphology of a hollow interior but an external continuous “cellular” structure, proves that the distribution of binary polymer PLA and CS in the island-like fibers.

Chemical Composition Analysis of the Shell Layer. In order to prove that the major component of the outer layer was CS, an XPS analysis was carried out. (Fig. 3A, B, C) show the XPS C1s spectrum of electrospun fibers: PLA, CS, and a PLA/CS 70:30 core-shell structure, respectively. For the PLA shown in (Fig. 3A), the C1s signals attributed to the C-C, C-O, and C=O bonds of PLA were located at 284.5 eV, 285.2 eV, and 298.5 eV. Compared to pure PLA,

14 Environment ACS Paragon Plus

Page 15 of 36

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

an obvious difference could be found in the PLA/CS 70:30 nanofibers (Fig. 3C), not only for the C1s signals of the C-C, C-O, C=O bonds, but also for the C-N bond at 288.6 eV, which belongs to CS. For the elemental content analysis (C, O, N), the N atomic percentage in the PLA/CS 70:30 sample was 5.1%, which was close to that of 5.9% observed for the pure CS sample. This illustrates that the major component on the outer surface of the PLA/CS nanofibers is CS. In comparison with the C1s spectrum of CS in (Fig. 3B), the shift of the C-N bond in the PLA/CS blended fibers from 287.6 eV to 288.6 eV might be caused by the interaction between the PLA and CS molecules. In addition, the whole survey is also displayed in (Fig. 3D) for the electrospun PLA/CS nanofibers. In comparison with (Fig. 3D c and d), a clear N1s signal disappeared after the fibers were rinsed with deionized water. Hence the results further confirm that the main component of the core layer was PLA, while CS mostly existed in the shell layer. The reason for this behavior might be that the positively charged amino groups of CS tended to move out due to electrostatic repulsion effects.

15 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

B

c-c

c-o c=o 292

290

288

286

284

282

intensity (cps)

Intensity (cps)

A

C-O C-C C-N

291 290 289 288 287 286 285 284 283 282 281

Binding Energy (eV)

Binding Energy (eV)

C

D

C-O C-C

C-N C=O

Intensity (cps)

Intensity (cps)

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

Page 16 of 36

O1s C1s

d c

N1s b a

292

290

288

286

284

Binding Energy (eV)

282

1200

1000

800

600

400

200

Binding Energy (eV)

Figure 3. XPS analysis of electrospun nanofibrous membranes: A pure PLA; B neat CS; C PLA/CS 70:30; D the survey scan of nanofibrous films; a PLA; b CS; c PLA/CS 70:30; d PLA/CS 70:30, after etching in deionized water for 24 h. The C-N bond at 288.6 eV contributing CS component demonstrated that the major component on the outer surface of the PLA/CS nanofibers is CS.

Laser Scanning Confocal Microscope Analysis of FITC Labelled nanofibers. To further confirm that the main component in the shell of the PLA/CS nanofibers was CS, FITC (fluorescein isothiocyanate)-labeled nanofibers were observed by LSCM. The N=C=S groups of the FITC fibers could interact with the -NH2 groups of CS. As shown in (Fig. S3), and as was

16 Environment ACS Paragon Plus

Page 17 of 36

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

expected, in the case of pure PLA nanofibers there was no fluorescence, while green fluorescence was visible on the surface of the PLA/CS composite sample. This indicates that some component on the surface of the PLA/CS nanofiber reacted with FITC, and it is therefore supporting evidence for CS being mainly present in the shell of the fibers. Crystallinity and Thermodynamic Analysis. The XRD and DSC curves of PLA/CS nanofibers synthesized using different blending ratios and electrospinning temperatures, demonstrated the changes in the crystallinity and thermodynamics of PLA. As shown in (Fig. 4A), raw PLA showed strong diffraction peaks at 16.7o and 19.0o, corresponding to the (200) / (110) and (203) crystal planes, and relative weak peaks at 14.9o and 22.3o. While there were no acute diffraction peaks in the PLA nanofibers, contrary to the raw PLA particles, the diffraction intensity of the blending fibers presented an obvious downtrend as the CS content increased. The DSC curve of CS is shown in (Fig. 4B). As it is known, CS is a crystalline polymer, and therefore it could not be melted, so its DSC curve showed a smooth line. However, the glass transition (Tg), cold crystallization (Tc), and melting temperatures (Tm) of PLA were clearly visible on the DSC curve of the PLA/CS nanofibers. From the data, it was clear that the Tg of PLA presented a slight tendency to increase with the proportion of CS. This could be interpreted by the partial miscibility of CS and PLA. In fact, the entanglement between the PLA and CS molecular chains in the blending system blocked the free motion of the PLA chains, which was accompanied by the migration of the CS molecular chains to the surface of the fiber, and the formation process of the core-shell structured nanofibers, as was observed in the microscopic phase separation shown in TEM micrographs. In addition, compared with the pure PLA nanofibers, it is worth mentioning that for all the PLA/CS blended systems, an obvious cold crystallization process appeared in the 100-160°C temperature range, while double melting peaks of PLA appeared at 160-180°C, especially for the PLA/CS 30:70 blended system, as is shown in (Fig. 4B d and e).

17 Environment ACS Paragon Plus

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

Page 18 of 36

These results confirmed that a significant amount of CS molecular chains interfered with the crystallization of PLA during the electrospinning process. As the SEM micrographs presented in (Fig. 1), the electrospinning temperature also strongly affected the fibers morphology. In order to corroborate our interpretation, it was worth investigating the effect of temperature on the crystallization and thermodynamic behavior of the blended fibers. The XRD pattern of PLA/CS 70:30 in (Fig. 4C c) shows that there was no significant differences in diffraction peaks at higher electrospinning temperature, but the melting temperature (Tm) of PLA obviously decreased. It was easy to understand this phenomenon because of the inhibition of crystallization by the rapid evaporation of the solvent evaporation at higher temperatures. The thermal properties of pure PLA and PLA/CS nanofibrous membranes are summarized in (Table S1), and are based on the DSC curves (Fig. 4B and D).

Turing Mechanism of Island and Core-shell Structure. To explain the formation mechanism of core-shell and island morphology, it was importent to investigate the ternary phase diagram of binary polymer-solvent system to track of the change of component proportion during the solvent evaporating. As shown in (Figure S4), accompanied by the solvent evaporation and cooling, electrospun solution into unstable or metastable region, phase separation occurred by nucleation and growth (NG, binodal phase separation) or by spinodal decomposition (SD, spinodal phase separation). In more detail, our previous work already explained the core-shell structure in PEO/CS electrospun blending system by this theory.32

18 Environment ACS Paragon Plus

f

10

20

30

40

50

e

d

c

b

a

B

a

exo

Intensity(a.u.)

A

b c d e f 40

60

60

80

100 120 140 160 180 200

Temperature (oC)

2 Theta (deg.)

c b

Heat Flow

exo

C

Intensity (a.u.)

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

Heat Flow

Page 19 of 36

D c

b

a

a 10

20

30

40

50

40

60

80

100 120 140 160 180 200 o

Temperature( C)

2Theta (deg.)

Figure 4. XRD (A) and DSC (B) curves of PLA/CS fibers with different blending ratios a: raw PLA b-f: fibers

b: PLA c: PLA/CS 70:30

d: 50:50

e: 30:70 f: CS

XRD curves (C) and DSC curves (D) of PLA/CS 70:30 fibers prepared at different temperature a: 25 °C

b: 45 °C

c: 60 °C. The Tg of PLA presented a slight tendency to increase with the

proportion of CS due to the partial miscibility of CS and PLA. All the PLA/CS blended systems, an obvious cold crystallization process appeared in the 100-160°C temperature range, confirmed that CS molecular chains interfered with the crystallization of PLA. The Tm of PLA obviously decreased with electrospun temperature.

19 Environment ACS Paragon Plus

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

Page 20 of 36

In addition, accompanied by the solvent evaporating, the charged groups of cationic CS molecular chains preferred to immigrate out under electrostatic repulsion was another crucial factor. There was an interesting phenomenon that why island structure appeared under higher electrospinning temperature, not core-shell structure? It was easy to understand that the evaporation rate of solvent was quicker under higher temperature, and there was not enough time to migrate to the surface of electrospun jets for some CS molecular chains, so that there was not enough CS component to cover the whole fibrous shell, and the intermittent islands structure appeared. Mineralization Analysis. The mineralizing behavior of the PLA/CS fibrous scaffold in a 10 × SBF solution for varying time (3, 6, 9, 12, and 24 h) was characterized by SEM, EDS, XRD, and TGA. As shown in (Fig. S5(1)), and in comparison with a pure PLA fibrous film, a significant number and size increase in the mineral crystals was observed on the surface of the PLA/CS 70:30 sample as the mineralization time increased. After mineralization for 9 h, a thin mineral layer formed on the surface of the fibers. After 24 h, the porous structure of fibrous scaffold had been fully covered by large mineral crystals and a highly roughened structural layer. The amount of deposited minerals gradually increased with the incubation time, and the weight percentages by TGA analysis, were 22.2%, 41.1%, 61.9%, and 68.2% as shown in (Fig. S5(2)). As shown in the XRD patterns, the characteristic diffraction peaks of the crystals, such as the (002), (210), (300), and (203) lattice planes at around 2θ = 26.0°, 27.6°, 31.7° and 45.5°, and the Ca/P=1.63 tested by EDS, well corresponded with the HA crystals. Conversely, only a few mineral particles were deposited on the pure PLA fibrous material after a 24 h incubation time. Previous studies39 have concluded that enhanced hydrophilic properties, which could be introduced through the use of certain functional groups, such as −OH and −NH2, could accelerate the deposition of calcium phosphate. This could enrich the amount of Ca2+ on the matrix surface,

20 Environment ACS Paragon Plus

Page 21 of 36

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

and lead to the local super saturation of Ca2+, and crystal nucleation. Therefore, our results confirmed that CS, which has an abundance of −OH and −NH2 groups, played a key role in the formation of hydroxyapatite on the surface of PLA/CS scaffolds. The porosity of electrospun fibrous membrane was measured using the following equations: Apparent density (g ⁄ cm3) = mat mass (g) /mat volume (cm3) Porosity = [1- PLA mat apparent density (g ⁄ cm3) / bulk density of PLA (g ⁄ cm3)] × 100% Pure PLA mat, the porosity is 85%±5%, while PLA/CS island-like mat 93±4%. Typical stress-strain curves of PLA and PLA/CS fibrous membranes better showed the different initial modulus of the membrane and tensile strength (Fig. S5(3)), and the tensile strength and stress(%) of the cross-linked fibrous membrane (PLA/CS 70:30) are 2.0 MPa and 20%, respectively. Adhesion and Growth of MC3T3-E1 Cells. The adhesion and proliferation of cells to the surface of biomaterials is a key step in tissue engineering applications. (Fig. 5) presents the CKK-8 test results, which revealed the proliferation of MC3T3-E1 cells seeded on the different scaffolds for 1, 2, 3, 7, and 14 days. It can be seen that, in comparison with the negative control, no statistically significant difference was detected in MC3T3-E1 cell numbers cultured for 3 to 14 days in the presence of PLA/CS nanowebs. In addition, the viability of the cells was above 95%. For pure PLA and surface CS island-like PLA scaffolds, statistically significant differences (p < 0.05) were observed after 2 seeding days. These results indicate that the CS island-like structure was beneficial for cell proliferation. Besides, visible increases of MC3T3-E1 cell numbers with the culture time are shown in (Fig. 6 and Fig. S6). After 3 days of culturing, the cell numbers on the island PLA/CS nanowebs were significantly higher than that of the other samples. This indicates that the combination of CS and rough nanosized morphologies played a

21 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

positive role on the attachment and growth of the cells. To more visually observe the spreading behaviour of MC3T3-E1 cells on the nanowebs, the cells were stained for 48 h with DAPI, for the nuclei of the cells, and TRITC-phalloidin for the cytoskeletal actin, in order to reflect the difference of cell attachment and cytoskeleton shape on pure PLA, PLA/CS with core-shell or island structure. As shown in (Fig. 7), the cells adherent amount and spreading morphology on PLA/CS nanowebs both of core-shell and island were obviously better than that on pure PLA sample, especially for island PLA/CS, indicating that this nanotopology surface was more favorable for the spread of cells. And SEM images with larger magnification provided more clear characterization of cellular spreading after culturing 96h, as shown in (Figure. S7).

4 Cell number ( × 10 )

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

Page 22 of 36

12

9

positive control PLA PLA/CS PLA/CS,mineralization negative control

*

*

* *

7 Day

14 Day

6 3

1 Day

2 Day

3 Day Time

Figure 5. CCK-8 assay of MC3T3-E1 cells in the extract of pure PLA, PLA/CS 70:30, and mineralized PLA/CS scaffolds after culturing for 1, 2, 3, 7 and 14 days (*p>0.05). In comparison with the negative control, no statistically significant difference culturing for 3 to 14 days in the presence of PLA/CS membranes. For pure PLA and surface CS island-like PLA scaffolds, statistically significant differences (p < 0.05) were observed after 2 seeding days.

22 Environment ACS Paragon Plus

Page 23 of 36

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

In a dynamic environment, cells respond to external stimuli by changing their function, or by reorganizing their cytoskeleton. Cells interact with the bioactive site on material surface through receptors on their surface, such as integrin. The interaction between biomaterials and cells is essentially to enhance the cell adhesion and growth on the material. Cell adhesion is the basis, as it will affect the cell proliferation, differentiation, and other functions. Cell adhesion on the material surface is divided into four steps: cell attachment, spreading, organization of actin cytoskeleton, and formation of focal adhesions. From a biomaterials viewpoint, the adhesion and growth of cells on the surface of a material strongly influenced by the surface properties of the biomaterial, such as the balance of hydrophilicity and hydrophobicity, surface roughness, and whether there are bioactive sites for anchoring the cells, etc. In this paper, in order to enhance the bioactivity of cells on the surface of scaffolds, CS was used to modify the surface of a PLA scaffold, so as to modify the abovementioned three parameters. For example, we provided active sites for integrin and cell adhesive proteins, and changed the scaffold surface from hydrophobic to hydrophilic, as many studies have demonstrated that the aforementioned cell behaviors, i.e. adhesion and growth, is more favorable on moderately hydrophilic interface than hydrophobic ones38. On the other hand, the roughness of the modified nanoscale surface and the surface topology of the scaffold were also significantly changed by the phase separation between PLA and CS. In general, the cells orient and migrate along the protrusions or grooves on the surface of a material. This phenomenon is called the contact induction of cell culture.

23 Environment ACS Paragon Plus

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

Page 24 of 36

Figure 6. LSCM micrographs (×400) of cells attached and grown on different scaffolds, stained with DAPI (the images in the first, second, and third rows corresponded to 1 day, 2 day, and 3 day, respectively) A: PLA (A1, A2, A3); B: PLA/CS 70:30 (B1, B2, B3) with core-shell structure; C: PLA/CS 70:30 (C1, C2, C3) with island-like structure. Visible increases of MC3T3-E1 cell numbers with the culture time are shown in (Fig. 6 and Fig. S6). After 3 days of culturing, the cell numbers on the island PLA/CS membrane were significantly higher than that of the other samples.

24 Environment ACS Paragon Plus

Page 25 of 36

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

A1

A2

B1

C1

C2

B2

Figure 7. SEM and LSCM micrographs (×400) of cells grown and attached for 48 h on different samples, stained with DAPI and TRITC-phalloidin.

A pure PLA (A1, A2); B PLA/CS 70:30

with core-shell structure (B1, B2); C PLA/CS 70:30 with island structure (C1, C2).

ALP Activity Test and Osteocalcin Expression of MC3T3-E1 on the Membrane. To evaluate the differentiation behavior of MC3T3-E1 cells on membrane with different surface components and protrusion structures, ALP activity and osteocalcin expression measurements are essential. ALP is an early marker for osteoblast differentiation, and osteocalcin is a typical marker of biomineralization of bone and mature osteoblasts. As shown in (Fig. 8), in comparison with the PLA scaffold, the ALP activities of cells cultured on the island-like and core-shell PLA/CS scaffolds were both enhance. In addition, at the day 7 and day 14 time points, the cells on the surface of fibers rich in CS showed a significantly higher expression of osteocalcin, especially the island-like PLA/CS scaffold. Early studies have noted that changes in the shape of

25 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

cells can affect the differentiation, and that a greater cytoskeletal tension of the cells tends to enhance their osteogenesis ability.39 Besides, the gene expression test including ALP, OPN and OCN of MC3T3-E1 cells during the 21 days. As shown in (Figure. 9). Compared to pure PLA membrane, there was a significantly higher ALP, OPN and OCN gene expression on PLA/CS membrane both 7 days and 21 days (*p < 0.05). After 7 days culturing, ALP gene expression increased slowly, but OCN expression significantly increased. Compared two kinds of PLA/CS fiber surface, core-shell and island-like structure, at early time 7 days, PLA/CS island-like membrane showed a higher ALP and OPN gene expression, while higher OCN gene expression at 21 days. All the data just supported our LSCM results of cells spread on scaffolds (Fig. 7). The cells grown on island-like scaffolds showed elongated morphologies and enhanced cytoskeletal tension.

0.6

(a)

PLA PLA/CS core-shell

1.2

*

0.5

PLA/CS island

1.0

Osteocalcin

1.4

ALP Activity absorbance at 405nm

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

Page 26 of 36

* 0.8 0.6 *

0.4 0.2

0.4

PLA PLA/CS core-shell PLA/CS island *

*

(b)

0.3 0.2 0.1

*

0.0

0.0 Day 3

Day 7

Day 14

Day 21

Day 7

Day 14

Figure 8. (a) ALP activity of MC3T3-E1 cells grown on the nanofibrous membranes cultured for 3, 7, 14 and 21days; (b) Osteocalcin expression of MC3T3-E1 cells on membranes cultured for 7 and 14 days; (*) denotes significant difference as compared to control (*P