Accepted Manuscript Incorporation of aligned PCL-PEG nanofibers into porous chitosan scaffolds improved the orientation of collagen fibers in regenerated periodontium Wenlu Jiang, Long Li, Ding Zhang, Shishu Huang, Zheng Jing, Yeke Wu, Zhihe Zhao, Lixing Zhao, Shaobing Zhou PII: DOI: Reference:

S1742-7061(15)30021-0 http://dx.doi.org/10.1016/j.actbio.2015.07.023 ACTBIO 3791

To appear in:

Acta Biomaterialia

Received Date: Revised Date: Accepted Date:

22 February 2015 11 July 2015 14 July 2015

Please cite this article as: Jiang, W., Li, L., Zhang, D., Huang, S., Jing, Z., Wu, Y., Zhao, Z., Zhao, L., Zhou, S., Incorporation of aligned PCL-PEG nanofibers into porous chitosan scaffolds improved the orientation of collagen fibers in regenerated periodontium, Acta Biomaterialia (2015), doi: http://dx.doi.org/10.1016/j.actbio.2015.07.023

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Incorporation of aligned PCL-PEG nanofibers into porous chitosan scaffolds improved the orientation of collagen fibers in regenerated periodontium

Wenlu Jianga, Long Lib, Ding Zhangc, Shishu Huangd, Zheng Jinga, Yeke Wue, Zhihe Zhaoa*, Lixing Zhaoa*, Shaobing Zhoub** a

State Key Laboratory of Oral Diseases, Department of Orthodontics, West China School

of Stomatolgy, Sichuan University, Chengdu 610041, PR China b

Key Laboratory of Advanced Technologies of Material, Minister of Education, School of

Materials Science and Engineering, Southwest Jiaotong University, Chengdu 610031, PR China c

Department of Stomatology, Peking Union Medical College Hospital, Beijing 100730, PR

China d

State Key Laboratory of Oral Diseases, West China School of Stomatolgy, Sichuan

University, Chengdu 610041, PR China e

Department of Stomatology, Affiliated Hospital of Chengdu University of TCM, Chengdu

610075, PR China

* **

Corresponding authors. Tel.: +86 28 85501442; fax: +86 28 85582167. Corresponding author. Tel.: +86 28 87634068; fax: +86 28 87634649.

E-mail addresses: [email protected] (Z. Zhao), [email protected] (L. Zhao), [email protected] (S. Zhou). 1

Abstract The periodontal ligament (PDL) is a group of highly aligned and organized connective tissue fibers that intervenes between the root surface and the alveolar bone. The unique architecture is essential for the specific physiological functionalities of periodontium. The regeneration of periodontium has been extensively studied by researchers, but very few of them pay attention to the alignment of PDL fibers as well as its functionalities. In this study, we fabricated a three-dimensional multilayered scaffold by embedding highly aligned biodegradable poly (ε-caprolactone)-poly (ethylene glycol) (PCE) copolymer electrospun nanofibrous mats into porous chitosan (CHI) to provide topographic cues and guide the oriented regeneration of periodontal tissue. In vitro, compared with random group and porous control, aligned nanofibers embedded scaffold could guide oriented arrangement and elongation of cells with promoted infiltration, viability and increased periodontal ligament-related genes expression. In vivo, aligned nanofibers embedded scaffold showed more organized arrangement of regenerated PDL nearly perpendicular against the root surface with more extensive formation of mature collagen fibers than random group and porous control. Moreover, higher expression level of periostin and more significant formation of tooth-supporting mineralized tissue were presented in the regenerated periodontium of aligned scaffold group. Incorporation of aligned PCE nanofibers into porous CHI proved to be applicable for oriented regeneration of periodontium, which might be further utilized in regeneration of a wide variety of human tissues with a specialized direction. Keywords: Periodontal tissue engineering; Biomimetics; Electrospun scaffold; Oriented regeneration; Periodontal ligament

2

1. Introduction Periodontal disease, usually due to trauma or infection, is characterized by destruction of the supporting structures of the teeth, including the periodontal ligament (PDL), alveolar bone and cementum [1]. It is one of the most prevalent diseases in humans and the main pathogenic cause of adult tooth loss, affecting oral function and quality of life [2]. Therefore, treatment for periodontal disease is widely concerned both in clinical and basic research. The ultimate goal of periodontal therapy is the regeneration of the original architecture and function of the periodontal complex, which involves the formation of new cementum on the tooth root, along with new periodontal attachment between newly formed bone and cementum [3]. This outcome is unpredictable with conventional therapy (i.e. GTR, Guided Tissue Regeneration) [4], largely due to the structural complexity of the periodontium, which consists of both the soft (PDL) and the hard (bone, cementum) tissues. To acquire the functionalized healing of periodontal tissue, the attachment between soft and hard tissue interfaces with oriented PDL fibers is critical in the regeneration process. The periodontal tissue relies on a highly aligned and organized microstructure of PDL in order to impart their desired physiological functionalities. Under physiological conditions, the PDL fibers are organized almost perpendicular to the tooth surface, and their ends are embedded in cementum and alveolar bone [5]. Such unique architectures fix tooth in the socket, providing support and protection for tooth under masticatory force and keep the structural integrity of the periodontium [6]. Experiments in animal models and patients have shown that when such arrangement of PDL are displaced, altered or removed by mechanical means or biochemical irritations, the amorphous type of tooth support usually leads to loss of function and resorption or ankylosis of the root [7-9]. Furthermore, the ageing of periodontium characterized by reduction 3

of occlusal force is also marked by structural and orientation alterations of PDL, even in those unaffected by periodontal disease [10]. It has also been observed that when loads are applied to teeth without proper intervening PDL, only a limited amount of bone remodeling occurs [11]. Over the years, most researchers have been attempting to regenerate the alveolar bone, yet the outcomes indicated no obvious restore of PDL attachment or cementum [12, 13]. In studies that paid attention to PDL regeneration, only a few observed newly formed PDL-like tissue with proper orientation [14, 15], most studies showed amorphous soft tissue encapsulated or parallel to the root surface [16-18]. Recently, a biomimetic 3-D rapid prototyped scaffold was developed with aligned internal pores to guide the direction of regenerated tissues, by which properly oriented PDL-like bundles was formed in vivo [19]. However, its micro-scale structure could be less effective in inducing elongated cellular morphologies and regulating corresponding cellular behaviors when observed at the submicro level. Therefore, it may be hypothesized that a scaffold which provides a more precise topographic guidance for each individual cell could be better utilized in the functionally oriented regeneration of periodontium. Meanwhile, it is widely understood that aligned nanostructure can induce aligned morphologies in most cell types through contact guidance [20-23], which are especially important in designing tissue engineering environments that mimic aligned tissues such as ligament and tendon. In addition, aligned cell morphology generally correspond to changes in cytoskeletal arrangement that can further trigger variations in other cell behaviors such as elongation, proliferation, migration, ECM production, and differentiation into cellular phenotypes that exhibit an elongated morphology [24,25]. Composed of tendon-like collagen bundles that are capable of bearing occlusal forces, the PDL can also be considered as a specific type of tendon/ligament [26]. We postulated that a

4

fiber-guiding nanoscaffold with aligned nanotopology could better facilitate and guide the oriented regeneration of PDL than micro-level scaffold with oriented pores. In this study, we fabricated a three-dimensional aligned layer-by-layer nanofiberous scaffold by incorporation of PCL-PEG (PCE) nanofibers into porous chitosan (CHI) to dictate the oriented regeneration of periodontium with aligned PDL fibers perpendicular to the root surface. Topographic-guiding and periodontium healing potentials of this scaffold were further evaluated both in vitro and in vivo.

2. Materials and methods 2.1. Scaffold preparation 2.1.1 Electrospinning of PCE nanofiber mats ε-caprolactone (ε-CL) and polyethylene glycol (PEG, Mn = 6 kDa) were purchased from Sigma-Aldrich Co. LLC. (St. Louis, MO, USA). All other chemicals and solvents were purchased from Kelong Chemical Reagents (Chengdu, China) with reagent grade or better, without further purification. PCE (PCL:PEG = 90:10, w/w) was synthesized by ring-opening polymerization (ROP) of ε-CL initiated by PEG and Sn(Oct )2 as a catalyzer [27]. The electrospinning process of the PCE nanofibers was followed by our previous report [28]. In brief, 2 g PCE was dissolved in 7.5 mL DCM and 2.5 mL N, N- Dimethyl Formamide (DMF) to form 20 wt% electrospinning solution. Electrospinning was carried out under 21 ± 0.5 kV applied voltage, at 17 cm tip-to-collector distance and 1 ml/h feeding rate with a 23 G stainless steel needles. The random oriented and aligned electrospun nanofibers were collected by stainless drum at 20 rpm and 1800 rpm respectively. All scaffolds were then dried under vacuum at room

5

temperature over 1 week for completely removing the residual solvents. Subsequently, the mats were used for further fabrication of PCE nanofiber embedded chitosan-based scaffolds. 2.1.2. Preparation of electrospun nanofiber embedded chitosan-based scaffolds The PCE nanofiber embedded chitosan-based scaffolds were prepared as showed in Fig. 1. Chitosan was purchased from Sigma-Aldrich Co. LLC. (St. Louis, MO, USA). CHI were dissolved in acetic acid aquious solution (2 %, w/v) (Kelong Chemical Reagents, Chengdu, China) by stirring for 12 h at 40 ℃. Genipin crosslinker (ConBon Bio-tech Co., Ltd., Chengdu, China) was added to the solution at 0.5% w/w with respect to the CHI amount. Electrospun PCE nanofiber mats were cut into 24×32 mm rectangular and stacked layer-by-layer into the prepared CHI-geniping mixture in an 8 cm2 culture dish. The dish was then sonicated for 10 minutes to generate complete infiltration of PCE nanofibers with CHI-geniping mixtrue. A cover clip was placed on each stack with a glass ring on top for continuous pressure. The mixture was kept at room temperature until gel formed (approximately 24 hours), then pre-freezed at −20 ◦C for 12 h and freeze-dried 24 h to obtain nanofiber embedded porous scaffolds. After freeze-drying, samples were washed in 70%, 90 % and 100 % w% ethanol (Kelong Chemical Reagents, Chengdu, China), for 20 min to neutralize the acid content and then repeatedly washed in deionized water till pH of washing medium was 7. Washing was also aimed to remove un-reacted Genipin residual. Each scaffold has about 30 layers of nanofibers embedded within the porous CHI, with thickness around 4mm. In the following texts, random or aligned PCE nanofiber embedded chitosan-based scaffolds were defined as 3D-RD or 3D-AL. Random or aligned PCE nanofiberous mats were

6

defined as 2D-RD or 2D-AL. Freeze-dried porous chitosan without incorporated PCE mats served as Porous control scaffold. 2.2. Scaffold characterization 2.2.1 Morphology observation by scanning electron microscopy (SEM) To observe the cross-sections of the 3D scaffolds (3D-RD, 3D-AL and Porous CHI scaffold), they were immersed in liquid nitrogen for 10 min and sectioned with a sharp razor blade. The newly exposed sections were coated with a 30 nm thick platinum for further observation. To observe the morphology of each internal layer in 3D-RD and 3D-AL multilayered scaffolds, the embedded nanofiberous layers was mechanically peeled off with tweezers. The peeled layers of 3D scaffold, together with 2D mats (2D-RD, 2D-AL) were also coated with a 30 nm thick platinum. All coated samples were observed under an Inspect™ F50 scanning electron microscope (FEI, Eindhoven, The Netherlands). The distributions of fiber diameter and angles of fiber orientation were calculated using the Image-Pro Plus 6.0 software (Media Cybernetics, Silver Spring, MD, USA). 2.2.2 Adhesion test to determine internal cohesion The cohesion between nanofibrous layers of the multilayered scaffold was evaluated by developing an adhesion test procedure in air at room temperature. Compression plates were used and fitted on an Instron 5567 universal test machine (Instron, Canton, MA, USA). Double-sided tape was placed on the surface of the compression plates and coated with a thin layer of superglue for fixing the specimen. The test specimen was cut to 1×2 cm from the top view, then placed between the plates and compressed up to -30 N and allowed to sit for 1 min to allow the glue to dry. The adhesion test consisted of a classic tensile test was performed at a stretch rate of 7

13 mm per minute until two layers were separated or "peeled off". The data acquisition ratio was set to 20.0 Hz. The real time displacements and corresponding loads were recorded. The internal cohesion of porous CHI scaffold was also tested following the same experimental procedure. 2.3. In vitro evaluation of scaffolds culturing with rat bone marrow mesenchymal stem cells (rBMSCs) 2.3.1 Cell culture and scaffolds preparation All reagents were purchased from HyClone Co. Ltd. (Logan, UT, USA). rBMSCs were obtained from 2-week-old newborn mice by whole bone marrow culture method and cultured in α-modified Eagles medium (α-MEM) supplemented with 10% FBS,1% penicillin-streptomycin, and incubated in 5% CO2 at 37°C. The cells from passage 3 to 5 were used. For evaluation of cell morphology and genes expressions on 3D scaffolds in contrast with 2D mats, all scaffolds were cut into circular pieces with diameters of 1.5 cm in a 24-well-plate, with a density of 5×104 cells/piece planted on the surfaces; For cell infiltration and viability test, 3D multilayered (3D-AL, 3D-RD) and porous control scaffolds were cross-sectioned into blocks with thickness of approximately 2mm in a 12-well-plate, with a density of 105 cells/block planted on the cross sections； For in vivo evaluation of periodontium regeneration, 3D multilayered (3D-AL, 3D-RD) and porous control scaffolds were cross-sectioned into slices with thickness of approximately 0.5 mm. All scaffolds were pre-sterilized by ultraviolet irradiation through UV lamps for 4 h and 75% ethanol for more than 4 h, followed by washing three times with PBS before use. 2.3.2 Cell morphology on scaffolds using SEM and fluorescent microscopy Cells planted on surfaces of each scaffold were cultured for 7 days. For SEM observation of cell morphology and elongation, samples were washed twice with PBS and fixed with 2.5% 8

glutaraldehyde (Kermel Chemical Reagent Co., Ltd., Tianjin, China), first for 1 h at room temperature, then overnight at 4°C. In order to observe the cytoskeleton arrangement of cells, the samples were fixed with 4% paraformaldehyde (Boster Biological Technology Co., Ltd., Wuhan, China) for 30min, and then stained by Alexa Fluor® 488 phalloidin (Invitrogen Ltd, Paisley, UK) or F-actin and 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) (Roche Diagnostics GmbH, Mannheim, Germany) for nuclei, then observed by an Olympus IX-71 Inverted Fluorescent Microscope (Olympus, Tokyo, Japan). By determination of the long axis of the nuleus and actin fibers of each adhered cell, the angulations of cell alignment in the fluorescent microscopic figures were measured with ImageJ software (National Institutes of Health, Bethesda, MD, USA). Three random views of each group in high magnification were selected. 2.3.3 Observation of cell infiltration into scaffolds by confocal microscopy For observation of rBMSCs infiltrating inside 3D-AL, 3D-RD and porous control scaffolds, the samples was stained by 0.01% acridine orange (Amresco, Solon, OH, USA) for 5 min in the dark after 3 days in vitro culture, washed three times and scanned by a Leica DMIRE2 laser scanning confocal microscope (Leica, Wetzlar, Germany). To quantitatively evaluated the infiltration depths, the microscopic images were reconstructed with Imaris 7.0 software (Bitplane, St. Paul, MN, USA) and subjected to oblique slicer (extended thickness: 30µm) along the direction of layers. The average width of fluorescence for each slice was measured and defined as depth of infiltration (n=5). 2.3.4 Cell viability in scaffolds with CCK8 test

9

The samples were cultured in a standard growth medium for up to 13 days. Each well was filled with 1.5 mL medium and 150µL of CCK-8 solution (Beyotime Biotech, Nantong, China) was then added at 1, 4, 7, 10 and 13 days. The samples were then incubated at 37°C for 2 h. The medium in the wells was extracted for absorbance measurement at 450 nm on a Thermo Varioskan Flash microplate reader (Thermo Scientific, Winooski, VT, USA). Three wells per group were subjected to replicate testing at each time point. 2.3.5 Evaluation of periodontal ligament-related genes expression by quantitative real-time polymerase chain reaction (qRT-PCR) To evaluate the effect of aligned nanotopography on cell differentiation, the expression of ligament-related genes of rBMSCs cultured on the scaffolds on 1, 7 and 14 days of in vitro culture were evaluate. Total RNA was isolated using TRIzol reagents (Invitrogen, Karlsruhe, Germany), and further reverse transcribed to cDNA using miScript Reverse Transcription kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The reaction products were then amplified in a real-time PCR reaction with a SYBR Green reaction Kit (Qiagen, Hilden, Germany) in a ABI-7300 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instruction. The sequences for the specific primer pairs (Takara, Tokyo, Japan) were: β-actin forward ACGGTCAGGTCATCACTATCG, reverse GGCATAGAGGTCTTTACGGATG

(155

bp);

Postn

forward,

ACAACTCCGTGTCTTCGTGTATCG, reverse, ATCTCTCGGAATATGTGAATGGCA (108 bp);

S100a4

forward,

TGCAGCTTCGTCTGTCCTTCTC

CTTCCACAAATACTCAGGCAACG, (118

bp);

Bglap

reverse, forward,

GCATTCTGCCTCTCTGACCTGG, reverse, GCTCCAAGTCCATTGTTGAGGTAG (131 bp); Alpl forward, AACCTGACTGACCCTTCCCTCT, reverse, TCAATCCTGCCTCCTTCCACTA 10

(104

bp);

Col3a1

forward,

GCTTTCAGTTCAGCTATGGCAATC,

GCAGTGGTATGTAATGTTCTGGGAG

(113bp);

Col1a1

reverse, forward,

CTGCTGGCAAGAATGGCGA, reverse, GAAGCCACGATGACCCTTTATG(164bp). Each reaction was performed in a final volume of 10 µl containing 2 µl of the cDNA, 0.5 mM of each primer and 1x SYBR Green PCR Master mix. The amplification profile was: denaturation at 95°C for 15 min, followed by 40 cycles of 94°C for 15 s, 55°C for 30 s and 70°C for 30 s, in which fluorescence was acquired. Each real-time PCR was performed on at least 3 different experimental samples. The amplification efficiencies of primer pairs were validated to enable quantitative comparison of gene expression. Representative results are shown as target gene expression normalized to β-actin. The relative level of expression of each target gene was then calculated using the 2-∆∆Ct method. Error bars reflect one standard deviation from the mean of technical replicates 2.4 In vivo evaluation of scaffolds in rat periodontal defect model 2.4.1 Surgically created periodontal fenestration defect in rats Four groups of female Sprague-Dawley rats (n=5) were designed: (i) blank control (no scaffold); (ii) porous control (porous CHI scaffold); (iii) 3D-RD and (iv) 3D-AL. For each rat, the defect was created on one randomly selected side. The standardized fenestration defect (3×2×2 mm3 dimension）was generated using a dental burr, exposing the mesia-palatal root surface of the maxillary 1st molar tooth (m-M1) to the epical level. The cementum layer on the exposed root surface was removed and recessed to expose the dentin surface. The root surface on both groups was conditioned with 24% EDTA (Biora, Malmo, Sweden) for 2 minutes to remove the smear layer and obtain a surface devoid of organic debris and then rinsed with normal saline. 11

The scaffolds were placed in cross-sectioned slices, each of which against the root surfaces then stuffed with bovine-derived porous xenograft Bio-Oss® (Geistlich Sons Ltd, Wolhusen, Switzerland) to immobilize the scaffolds as well as to fill the alveolar defect. The surgical process is shown in Fig. 2A. Specifically, for aligned nanofibers embedded group (i.e. 3D-AL), to control the orientation upon transplantation during surgery to ensure nanofibers perpendicular to the root surface, the scaffold was sectioned perpendicular to the direction of nanofibers alignment, and was then placed with sectioned profile against root surface, as in Fig. 2B(c). With 8 weeks time point, the rats were anesthetized to death with 40% chloral hydrate (4 mL/kg) (Kelong Chemical Reagents, Chengdu, China), and the whole maxilla was carefully cut off, washed with distilled water, and immediately placed into 4% paraformaldehyde solution (Boster Biological Technology Co., Ltd., Wuhan, China) for further evaluations. The experimental protocol was approved by the Sichuan University Institutional Animal Care and Use Committee. All of the experiments followed the Experimental Animal Protection Act. 2.4.2 Histomorphometric analysis for regenerated PDL-like tissue and collagen fibers formation with hematoxylin-eosin (H&E) and picrosirius red staining After fixing in 4% paraformaldehyde for 1 day, the harvested maxillas were decalcified in 10% ethylene diaminetetraacetic acid (EDTA) (Kelong Chemical Reagents, Chengdu, China) for 4 weeks and embedded in paraffin blocks for histological sectioning. The randomly selected sections were stained with both H&E or picrosirius red stains and imaged with a Nikon ECLIPSE 80i general light microscope (Nikon, Kanagawa, Japan) and a Nikon ECLIPSE E400POL cross-polarized light microscope (Nikon, Tokyo, Japan), respectively. Three randomly selected views (high magnification) of H&E stained sections within the surgical area in each group were subjected to measurement for angulations of fiber bundles against the root surfaces 12

with ImageJ software. Three randomly selected regions of picrosirius red stained sections within the surgical area were captured (high magnification) and semi-quantified for collagen I/III ratios using the Image-Pro Plus 6.0 software. 2.4.3 Immunofluorescence to determine functional tissue formation Fluorescence staining to periostin was performed on paraffin sections with primary rabbit polyclonal antibodies diluted at 1:200 (Abcam, Cambridge, UK). Immunological reaction was visualized by sheep polyclonal secondary antibodies to rabbit conjugated to Texas Red (TR) diluted at 1:1000 (Abcam, Cambridge, UK). The sections were then treated with DAPI (Roche Diagnostics GmbH, Mannheim, Germany) and covered with glass coverslips. The stained slides were imaged using an Olympus IX-71 Fluorescent Microscope and subjected to florescent intensity evaluation with Image-Pro Plus 6.0 software. The results were shown using IOD (Integrated Optical Density) per area. 2.4.4 Mineralized tissue formation with micro-CT analysis The fixed rat maxillaries were scanned by a Siemens Inveon micro-CT (Siemens, Erlangen, Germany). Based on the HU greyscale level, mineralized tissue formation was reconstructed and quantified by Mimics 10.01 software (Materalise, Leuven, Belgium). Regions of interests for the assessment of mineralized tissue formation were created using the following criteria: horizontal region of entire defect and vertical epical level. Bone volume fracture was defined as the average volume ratios of mineralized tissue in the defect. Bone density was defined as the average HU value of mineralized tissue in the defect. Alveolar height was defined as the ratio of length from tooth apical to alveolar crest to length from tooth apical to the cemento-enamel junction (CEJ), as L1/L2 shown in Fig. 10A(e). 13

2.5 Statistical analysis All experiments were repeated at least three times and data were expressed as mean ± SD. Statistical comparisons were made using factorial analysis of variance (ANOVA) for comparing treatments from controls or between two groups at same time point. P < 0.05 was considered as statistically significant.

3.Results 3.1. Characterization of PCE nanofibers embedded multilayered scaffolds Nanofibrous topography of 2D-RD, 2D-AL, 3D-RD and 3D-AL scaffolds were observed using SEM scan. The surface morphologies of each scaffold was shown in Fig. 3A, with respective distributions of fiber diameter and angles of fiber orientation. The principal axis of the fibers in random scaffolds (2D-RD or 3D-RD) was oriented in multiple directions. The majority of fibers in 2D-AL group were almost aligned in a single direction, while 3D-AL group showed concentrated distribution of fiber angles as well, despite some apparently curved nanofibers. The average fiber diameters between scaffolds of the same dimensions were not significantly different (p＞0.05), whereas 3D scaffolds showed significantly (p