Nonmulberry Silk Braids Direct Terminal Osteocytic Differentiation

Mar 22, 2017 - Nonmulberry Silk Braids Direct Terminal Osteocytic Differentiation through Activation of Wnt-Signaling. Swati Midha, Shibu Chameettacha...
2 downloads 8 Views 2MB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Non-mulberry silk braids direct terminal osteocytic differentiation through activation of Wnt-signaling Swati Midha, Shibu Chameettachal, Emeli Dey, and Sourabh Ghosh ACS Biomater. Sci. Eng., Just Accepted Manuscript • DOI: 10.1021/acsbiomaterials.7b00006 • Publication Date (Web): 22 Mar 2017 Downloaded from http://pubs.acs.org on March 31, 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 Biomaterials Science & Engineering 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 38

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 Biomaterials Science & Engineering

Non-mulberry silk braids direct terminal osteocytic differentiation through activation of Wnt-signaling Swati Midha, Shibu Chameettachal, Emeli Dey, Sourabh Ghosh* Department of Textile Technology, IIT Delhi, Hauz Khas, New Delhi-110016, India

* Corresponding author

Dr. Sourabh Ghosh Department of Textile Technology Indian Institute of Technology Delhi New Delhi, India - 110016 Email: [email protected] Phone: 91-11-2659-1440

ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering

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 2 of 38

Abstract Silk-based natural polymers can regulate osteogenesis by mimicking the extracellular matrix of bone and facilitate mineralized deposition on their surface by cultured osteoprogenitors. However, terminal differentiation of these mineralizing osteoblasts into osteocytic phenotype is not yet demonstrated on silk. Therefore, in this study we test the hypothesis that flat braids of natively (non-regenerated) spun non-mulberry silk A.mylitta, possessing mechanical stiffness in the range of trabecular bone, can regulate osteocyte differentiation within their 3D microenvironment. We seeded human pre-osteoblasts onto these braids and cultured them under varied temperatures (33.5oC and 39oC), soluble factors (Dexamethasone, Ascorbic acid and β-glycerophosphate) and cytokines (TGF-β1). In particular, cell dendrites were conspicuously evident, confirming osteocyte differentiation, especially, in the presence of osteogenic factors and TGF-β1 expressing all characteristic osteocyte markers (podoplanin, DMP-1 and sclerostin). A.mylitta silk braids alone were sufficient to induce this differentiation, albeit only transiently. Therefore we believe that the combinatorial effect of A.mylitta silk (surface chemistry, braid rigidity, topography), osteogenic differentiation factors and TGF-β1 were critical in stabilizing mature osteocytic phenotype. Interestingly, Wnt signaling promoted osteocytic differentiation as evidenced by upregulated expression of β-catenin in the presence of growth factor cocktail. This study highlights the role of non-mulberry silk braids in regulating stable osteocytic differentiation. Future studies could benefit from this understanding of the signaling mechanisms associated with silk-based matrices in order to develop 3D in vitro bone model systems.

Keywords: non-mulberry silk, Wnt/ β-catenin signaling, osteogenic differentiation, textile braid, osteocyte

ACS Paragon Plus Environment

1

Page 3 of 38

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 Biomaterials Science & Engineering

1. Introduction Tailoring the 3D architecture, surface chemistry, amino acid sequences or secondary conformation of silk fibres has emerged as an effective strategy for the fabrication of biomimetic silk-based materials for bone tissue engineering.1 Silk fibroin, the primary component of silk fibres, comprises of unique amino acid sequences (GAGAGS and GAGAGY) in repeating units which enables the tight packaging of protein chains into anti-parallel β-sheets and renders strength and resilience to the structures, akin to native bone matrix. Of the two widely known varieties of silkworm silk, Bombyx mori (mulberry) and Antheraea mylitta (nonmulberry); the latter has furbished exceptional potential in bone regeneration and mineralization both in vitro and in vivo.2 This is mostly attributed to the presence of an additional ArgGly-Asp (RGD) peptide-motif in A.mylitta. However, as the sequence of A.mylitta is only partially known, the existence and exact role of this sequence is still debatable.3 Moreover, silk-based tissue engineering mostly relies upon tailoring scaffolds by re-constitution of dissolved silk fibres into various 3D morphologies. Since A.mylitta silk is difficult to dissolve using standard laboratory solvents,2 their application has largely been limited. Apart from the exceptional mechanical properties, silk fibroin protein is considered a potential matrix for bone tissue engineering due to its striking resemblance to collagen type I.3 As proof of concept, previously we used silk fibroin from B.mori silkworm to investigate the role of material properties (chemical, structural and mechanical) in regulating hydroxyapatite deposition on silk using simulated body fluid.4 Interestingly, we noticed that B.mori braids, constituting of native silk fibres, regulated hydroxyapatite deposition along the c-axis (002), oriented longitudinally parallel to silk fibres, typically found in native bone. This was a very significant finding as the mechanism was found to be exactly similar to biomineralization of collagen type I in bone. On the contrary, regenerated 3D porous silk scaffold lacked this potential, most likely due to the disruption of amino acid sequences as a result of dissolu-

ACS Paragon Plus Environment

2

ACS Biomaterials Science & Engineering

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 4 of 38

tion and re-constitution of original protein assembly.5 This was followed by a comparative in vivo investigation on the osteogenic potential of 3D porous B.mori with its non-mulberry counterpart A.mylitta in critical size calvarial defects of rats.2 Radiographic and histological evidence revealed complete osseo-integration of implanted defects within 6 months, albeit only in A.mylitta scaffolds. Based on these results, we hypothesized that natively spun fibres of non-mulberry silk A.mylitta will provide a suitable matrix for bone regeneration studies. In native bone, osteoblasts initiate the synthesis and deposition of osteoid matrix wherein these cells get entrapped and start transitioning into dendritic osteocytes, characterized by genotypic expression of dentin matrix protein-1 (Dmp-1) and Sclerostin (Sost). Existing strategies to establish such osteocyte models include the use of commercial cell lines in 2D monolayer or animal model testing. However, the signaling cues provided by 3D architectures (biochemical, topographical and mechanotransduction) that direct cellular differentiation towards a particular lineage cannot be replicated in conventional monolayers. Animal testing, on the other hand, is expensive, plagued by ethical concerns and often fail to correlate with human physiology.6 To circumvent these issues, 3D culture systems of polymers and proteins are becoming popular. Till date in vitro studies on silk reported mineralized matrix deposition using osteoblast cultures, but we could not find any report on complete cellular differentiation and transition into terminally differentiated osteocytes. Therefore, with an aim to develop a 3D bone model on A.mylitta braids ex vivo by reproducing the phenotypic and genotypic transition of pre-osteoblasts to dendritic osteocytes, akin to native bone, we will address the following key questions; (i) examining cellular responses towards precisely fabricated hierarchical complexities of such non-mulberry braided structures, (ii) role of substrate properties in regulating osteocytic differentiation on silk matrices? (iii) major signaling pathways directing osteogenic signaling on non-mulberry silk matrices?

ACS Paragon Plus Environment

3

Page 5 of 38

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 Biomaterials Science & Engineering

Osteogenic signaling on silk scaffolds has been poorly understood so far.3 Jung et al. (2013) studied gene expression of rat bone marrow cells on B.mori silk fibroin proteins which demonstrated suppressed Notch-activated genes while upregulating the expression of osteogenic markers, thus indicating a critical role of silk fibroin proteins in suppressing Notch signaling.7 Little is known about the role of silk in driving the key osteogenic signaling mechanisms. Wnt signaling, one of the main bone signaling pathways, is regulated via β-catenin, a transcription factor known to express in the presence of transforming growth factor-β1 (TGFβ1).8 TGF-β1, most abundant cytokine in human bone (200 mg/kg), plays a central role in bone matrix turn-over,9 storage of mineral and generation of hematopoietic cell lineage. Varying the concentration of TGF-β resulted in varied outcomes in in vitro and in vivo studies.10 This poor correlation between the existing literature on in vitro and in vivo based experiments further emphasizes on the inadequacies of the current in vitro 3D culture systems for bone tissue engineering and demands for a reliable, long-term sustainable in vitro model to accurately capture the physiology of native bone tissue microenvironment.11 Such 3D in vitro models will not only facilitate improved understanding for fabrication of advanced materials, but also help in investigating pathological signaling mechanisms and screening of drug candidates for therapeutic applications. Considering the applications of this in vitro model for bone differentiation, hFOBs 1.19, a pre-osteoblastic cell line with surface markers similar to osteogenic progenitors12 served as an excellent choice for the study. While the use of cell lines is often plagued with concerns regarding genetic manipulation and misrepresentation of human physiology, an interesting trait about this cell line is that it is immortalized but non-transformed, transfected stably with temperature sensitive SV40 T-antigen.13 Gene expression analysis showed that at 39°C, hFOBs expressed upregulated levels of bone-related markers; Cbfa1, parathyroid hormone receptor and osteocalcin. Differentiation of hFOBs into mature osteoblasts synthesizing

ACS Paragon Plus Environment

4

ACS Biomaterials Science & Engineering

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 38

mineralized matrix in 3D culture systems is known,14-16 but no study has so far reported the transition of these cells into osteocytic phenotype on any material. Keogh et al. (2010) cultured hFOBs on collagen-glycosaminoglycan scaffolds and demonstrated early evidence of mineralization only after 35 days in vitro.14 Therefore, the underlying mechanisms that trigger terminal differentiation of pre-osteoblast cells to osteocytes, especially as a function of the underlying substrate, still remains a question. In this study, we test the hypothesis that non-regenerated, natively spun A.mylitta silk fibres fabricated into a braided morphology can regulate extracellular matrix synthesis and osteocytic transition of cultured cells within the 3D microenvironment. The effect of biomechanical, topographical and biological cues were investigated by culturing human preosteoblast cells in the presence or absence of bone inducing factors and TGF-β1. Osteogenic differentiation into mature osteoblasts/osteocytic phenotype was investigated by real-time gene expression profiling from early osteoblasts to terminal osteocytes, monitoring cell morphology and metabolic activity and protein synthesis by confocal microscopy. To the best of our knowledge, this is the first study reported on non-mulberry silk braids that maps the entire process of bone differentiation from pre-osteoblasts to stable osteocytes on A.mylitta braids. Beyond the conventional scope of clinical bone graft substitutes, this non-mulberry silk braid has immense therapeutic potential for being used as a 3D in vitro model system of bone differentiation for screening the efficacy of drugs targeted towards bone disorders. 2. Materials and Methods: 2.1 Source of silk yarn - A. mylitta silk fibres of 60 Nm (16.7 tex) count were procured from Starling Silk Mills Pvt. Ltd., Malda, Bengal such that silk double yarn (obtained by twisting two single yarns together) was produced with a resulting count of 30 Nm (33.3 tex) to develop braided structures, so that the yarns could withstand forces during braiding process.

ACS Paragon Plus Environment

5

Page 7 of 38

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 Biomaterials Science & Engineering

2.1.1 Fabrication of A.mylitta braids - Flat braids of silk were produced by passing a fixed number of yarns in a diagonally opposite direction to the subsequent layer such that each bundle alternates over and under another bundle of yarns. Then the structure is oriented at an angle to the circumferential axis of the resulting braid on a flat braiding machine with 34 two plied yarns. Figure 1A schematically illustrates the structural hierarchy of the flat braid used in the experiment and establishes the nomenclature used throughout the manuscript. The braids were subjected to degumming by boiling them in 0.02 M Na2CO3 solution for 30 min in order to remove sericin and subsequently dried overnight.17 Next day, dried braid fabrics were manually cut into scaffolds of smaller dimensions (4 mm×4 mm) for cell culture.4 2.1.2 Physical characterisation of braids. Fibre/Yarn Diameter - The diameter (d) of each respective fibre and yarn (n=150 each) in the A.mylitta braid was measured using projection microscope (WeswoxOptik microscope, MP-385A, India) on 40× and 10× magnifications respectively. Braid angle measurement - Braid angle is the angle between the tow (bunch of yarns) direction and the axis of the tube. Braid angle was measured from macroscopic images obtained using inverted optical microscope (Leica DM2500P, Germany) equipped with high resolution digital microscope camera system (Leica DFC425C, Germany). A total of 12 observations were taken across random regions in the braided sample. Packing Fraction - Packing fraction is the fraction of yarn volume in a fabric structure that is occupied by a collective bunch of fibres. Packing fraction of silk scaffolds was measured as described elsewhere.18 Packing fraction = total cross sectional area of the fibres / cross sectional area of yarn = Volume of fibre in yarn/Volume of yarn

ACS Paragon Plus Environment

6

ACS Biomaterials Science & Engineering

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 8 of 38

Twist per inch (tpi) - For measuring the twist density, untwist retwist method was used on Eureka Precision Instrument EY06 type. Yarn length of 20 cm (n=5) was used to measure the tpi. Tensile Strength - The tensile tests of A.mylitta braids (n=3) were carried out by H5KS Tensile Strength Tester materials testing machine under ambient conditions, i.e. 28°C and 70±5% relative humidity. Gauge length used was 75 mm with 300 mm/min cross-head speed. Compression - The compression test was performed using a ZwickRoell LTM 1000 machine on braids (n=3) measuring 10 mm×4mm. For measuring the silk braid compression, the Essdial Thickness Gauge was used where different levels of load from 50 gf/cm2 to 2000 gf/cm2 were applied and the consequential change in material thickness was recorded. 2.2 In vitro cell culture 2.2.1 Cell culture - Human fetal osteoblast cell line (hFOB 1.19 ATCC, CRL-11372) was purchased from Promochem, Bangalore, India. According to ATCC guidelines, hFOBs cultured at 33.5°C undergo rapid cell division whereas cell differentiation is observed at the temperature of 39°C.13 The cells were expanded in standard medium comprising of Dulbecco’s modified eagles medium/nutrient mixture F-12 Ham, 1:1 mixture (Cat No. AL187A, Himedia) with 10% FBS (Biological Industries, Cat No. 04-121-1A) and geneticin 3 µg/ml (Cat No. 10131035, Invitrogen) at 33.5°C in a 95% humidity and 5% CO2. 2.2.2 Cell seeding on braids – Scaffolds (4 mm×4 mm × 0.6 mm) were sterilized in 70% ethanol followed by pre-wetting in complete medium overnight followed by cell seeding with 1.72×105 cells per cm2 of the braided construct. After 2 days of incubation in standard media, half of the cell-seeded constructs were shifted to differentiation medium. Differentiation media contained standard media with 10 nM dexamethasone (Sigma-Aldrich, USA), 0.01 M βglycerol phosphate (Sigma-Aldrich, USA), 50 µg/ml ascorbic acid-2-phosphate (Sigma-

ACS Paragon Plus Environment

7

Page 9 of 38

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 Biomaterials Science & Engineering

Aldrich, USA) to promote cell differentiation. Media was changed twice a week. All cell culture condition sets were carried out at both 33.5°C and 39°C. 2.3 Analysing cell behaviour. 2.3.1 Quantification of cell attachment on braids - The braids (n=3) seeded with 2×105 hFOBs were placed in a 24 well plate with wells pre-coated with 2% Poly (2-hydroxyethyl methacrylate) (Sigma, USA, Cat No. P3932) in order to avoid cell attachment to the wells. Non-attached cells were collected by centrifugation, stained with trypan blue and counted using neubauer chamber to determine % adhesion. 2.3.2 Determining metabolic activity - The cellular metabolic activity was determined using standard MTT assay. After 1 and 7 days, hFOB-laden braids [(n=3), experiment repeated twice] were stained with tetrazolium MTT salt (1:10 ratio in media) at 37°C for 4 h followed by dissolution using dimethylsulfoxide. The absorbance was measured at 560 nm, using iMark microplate Absorbance reader (Biorad). Cell-free braids served as a control for this procedure. 2.3.3 Scanning electron microscopy - After 7 days, cell-laden braids were fixed, dehydrated with gradient alcohol series, gold coated (EMITECH K550×, UK) and visualized using SEM (model EVO 50, Zeiss, UK). To compute mean pore size, 20 pores were measured across 3 images per sample using Image J software (NIH, USA). 2.4.4 Real Time-Polymerase chain reaction (RT-PCR) - For the experiment, the total RNA was isolated at day 1, 7 and 14 from cells cultured on braids using RNeasy mini kit (Qiagen) as per manufacturer's protocol. RNA concentration was determined using Nanodrop 2000C (Thermo Scientific, Wilmington, USA) spectrophotometer and cDNA was synthesized using first strand cDNA Synthesis Kit (ThermoScientific, Cat No. K1612). Real time quantitative

ACS Paragon Plus Environment

8

ACS Biomaterials Science & Engineering

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 10 of 38

PCR was conducted using SYBR Green Master Mix (Quantitect, Cat No. 204074) and Rotor gene Q thermocycler (Qiagen). QuantiTect primers (Qiagen) used for gene expression analysis included alkaline phosphatase (ALP; Cat No. QT00211582), collagen type 1 alpha (COL1A1; Cat No. QT00037793), Runt-related transcription factor (RUNX2; (Cat No. QT00020517), Osteonectin (SPARC; Cat No. QT00018620), Osteopontin (SPP1; Cat No. QT01008798), Osteocalcin (BGLAP; Cat No. QT00232771), podoplanin (Cat No. QT01015084), DMP1 (Cat No. QT00022078), SOST (Cat No. QT00219968) and β-catenin (Cat. No. QT00077882). Glyceraldehyde-3-phosphate-dehydrogenase (GAPDH, Cat No. QT00079247) was used as the house keeping gene. hFOBs cultured in the form of 2D monolayer served as controls. The reactions were set up in triplicate with the whole experiment repeated twice. 2.4.5 Immunofluorescence studies – For immunofluorescence analysis, hFOB-laden braids were harvested at day 7, fixed, permeabilized and blocked.17 Staining was performed using anti-osteopontin (10 µg/ml, Millipore) followed by secondary staining with goat anti-mouse IgG antibody-FITC conjugate (1:200, Millipore) for 1 h at RT. Actin staining was performed for 30 min at RT using rhodamine phalloidin (Sigma, Cat No. P1951), followed by Alexafluor 546 (Cat No. A11003, Millipore, MA, USA). Nuclear staining was performed with DAPI (Sigma Aldrich, USA, Cat No. 32670). For image capturing, Leica TCS SP5 (Leica Microsystems) inverted confocal laser scanning microscope was used. 2.5 Statistical analysis - Data represented as the mean ± standard deviation, where n is the number of experimental repeats conducted. To determine statistical significance of data, Student’s t-test was conducted and probability values in the range of p < 0.05 were noted as significant. 3. Results

ACS Paragon Plus Environment

9

Page 11 of 38

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 Biomaterials Science & Engineering

3.1 Structural hierarchy of braids - The tpi value of the yarn was carefully optimised to 6.12 in order to achieve the desired mechanical properties required to withstand braiding and aid in compaction of the braided structure. The physical characterization of the fibres, yarn (Table I) and braids (Table II) was performed. Moreover, the braid depicted a flattened morphology comprising of several layers of hierarchy (Figure 1B). The measured thickness of the total braid was 0.61 mm. The criss-cross braid structure (Figure 1C) comprised of two plied yarns (Figure 1D), with 34 yarns intertwined to make a single layer of the braided scaffold. Each yarn (two plied) was further made up of 66 fibres; building blocks of the 3D braided architecture (Figure 1E). Quantitative analysis revealed a network of uniformly oriented multi-layer stack of fibres with a thickness of 21.9 µm + 0.2 µm each. Further to this, considering the dimensions of a human osteon (223 µm),19 we decided to use A.mylitta fibres in the 3D braided structure with average pore sizes close to this value (i.e., 196 µm) as determined with SEM micrographs. In addition, a constant braiding angle in the range of 35-40° with respect to the subsequent layer (Figure 1A) resulted in compaction of the structure due to the packing of large number of such fibres which was possibly a major reason for the resulting lower packing fraction and pore size of the braided fabric. Moreover, some silk fibres appeared loose in the textile braid structure possibly due to the degree of physical wear during the braid fabrication and/or degumming process (Figure 1B). This made the material surface relatively rough. The surface roughness of yarn as evaluated by AFM (Rq) corresponded to 77 nm (Figure 2). This increased surface roughness of the braids due to the interconnected porous structures aids in the formation of neo-bone tissue, allowing adequate integration of host tissue/cells and channels for nutrient/metabolic waste diffusion.20 Moreover, it is well known that increasing roughness and stiffness of scaffolds leads to superior differentiation of osteogenic progenitors.21 Particularly, human bone marrow stem cells differentiating into mature osteoblasts on stiffer matrices has been extensively researched.3 Also, the fact that braid

ACS Paragon Plus Environment

10

ACS Biomaterials Science & Engineering

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 38

was fabricated directly from native silk fibres without the usual dissolution process which further contributed towards the increased stability of the protein chains and resultant rigidity of the braided construct (Table II). Table I. Characterization of A.mylitta two plied yarn Yarn diameter (µm)

274

Linear density (tex)

33.33

Packing fraction (%)

42

Elongation (%)

9.6

Table II. Mechanical properties of A.mylitta braid Tensile Testing Ultimate tensile stress (MPa)

42.86

Ultimate tensile strain (%)

30.32

Initial modulus (N/tex)

15.88

Elongation (%)

9.6

Compression testing Compression % at 50 gmf

17.91

Young modulus (MPa)

1.41

3.2 Cell behaviour on braids 3.2.1 Initial cell attachment - Cell counting data revealed that approximately 59000 + 23 out of the total 2×105 hFOBs seeded were found floating/dead in culture wells on day 1, resulting in 70.5 + 6.7% of cell attachment (Table III). Table III. Summary of quantitation data from cell adhesion Matrix

hFOBs seeded

No. of floating cells

% Floating cells

No. of Attached cells

% Attachment

Braid (3D)

2 x 105

5.9 x 104

29.5

1.41 x 105

70.5

ACS Paragon Plus Environment

11

Page 13 of 38

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 Biomaterials Science & Engineering

Petri dish (2D)

2 x 105

1.6 x 104

8

1.84 x 105

92

3.2.2 Cell growth kinetics - Surprisingly at 33.5°C; proliferation temperature of cell line,13 no significant increase in metabolic activity of hFOB-laden braids was observed (p < 0.05), both with and without differentiation factors (Figure 3). However, when the temperature was raised to 39°C (differentiation temperature),13 metabolic activity of hFOBs declined by 2.4 fold (p < 0.05) in the absence of osteogenic differentiation factors, while the value increased significantly following treatment with osteogenic differentiation factors. This increase in hFOB activity at 39°C, which has not been witnessed earlier, emphasizes on the interaction of the osteogenic differentiation factors with the hFOB-seeded A.mylitta braids. 3.2.3 Morphological analysis - By day 1, conspicuous sheath of cells along with extracellular matrix components (which starts synthesizing as early as 4 h post cell seeding22 covered majority of the braided surface, encapsulating the underlying pores and inter-yarn spaces (Figure 4A), as cell attachment was significantly high (Table III). On higher magnification (2500×), the individual cells adhered onto the braids (Figure 4B; white arrow) and attained an elongated morphology along the long axis of the fibres (Figure 4C). This was possible as the diameter of the silk fibers was similar to that of hFOBs and hence ensured proper wrapping of the cells across the surface of the fibres in a 3D orientation. After 7 days of incubation at 33.5°C without the presence of any differentiation factors (Figure 4 D-F), a confluent sheath of cellular matrix was evident (Figure 4D). In regions where the cells were directly in contact with the braid, cell morphology appeared more rounded (Figure 4E, boxed regions), out of which few assumed long dendritic processes resembling osteocytic-like morphology (Figure 4E, inset) suggesting the role of A.mylitta surface in accelerating differentiation. On the other hand, hFOBs associated with the cell-synthesized matrix were more flattened (Figure 4F). In the presence of osteogenic factors, the cellular ma-

ACS Paragon Plus Environment

12

ACS Biomaterials Science & Engineering

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 38

trix was more prominent (Figure 4G). Due to this extensively spread sheath on the braided surface, most of the cells were entrapped within the matrix region spread across multiple fibres (Figure 4 H&I). When cultured at 39°C, a confluent cell sheath, observed in all other groups, was not evident in hFOB-laden braids cultured in the absence of differentiation factors (Figure 4J), also evident from lower MTT values (Figure 3). Cell morphology was highly irregular and distorted (Figure 4K, asterisk) with traces of debri/fragments (Figure 4L, arrowheads), which were possibly an outcome of cellular degradation due to apoptosis. On the contrary, in the presence of differentiation factors, a continuous sheath of cellular matrix (data not shown) invaded by distinct cellular morphologies indicated later different stages of osteogenic differentiation (Figure 4 M-O); (i) a rounded hFOB with high contrast particle deposition on the surface (asterisk ‘*’) indicative of mineral-like deposition (Figure 4M), (ii) typical osteocyte-like morphology with prominent dendritic protrusions (Figure 4 N&O) spread across multiple fibres (Figure 4O). This co-existence of osteoblasts and osteocytes on A.mylitta matrix may facilitate a bidirectional communication between the two cell types and generate important signals which may influence the differentiation of osteoblasts towards osteocytes,23 however, this is only speculation at this stage and would require controlled series of experimentation to be validated. 3.2.4 Effect of varying differentiation conditions on osteogenic gene expression - The expression of osteogenic-specific markers for hFOBs cultured either on braids (Figure 5) or 2D monolayer (Supplementary Figure S1) with and without differentiation factors at 39°C only. hFOBs cultured at 33.5°C (which is the proliferative temperature), demonstrated nominal expression of all differentiation markers throughout the culture period (data not shown).

ACS Paragon Plus Environment

13

Page 15 of 38

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 Biomaterials Science & Engineering

3.2.4.1 Osteoblast markers - The expression of ALP in hFOB-laden braids was increased with time in all the conditions, corresponding to 1.1 fold upregulation with differentiation factors and 30.6 fold upregulation without factors from 7 to 14 days (Figure 5). The addition of differentiation factors rapidly upregulated the expression by 6.4 fold as early as day 1 (p < 0.05). Addition of TGF-β1, drastically down regulated (p < 0.05) the ALP transcript levels at all time points in hFOB-laden braids (Figure 5) as well as 2D monolayer (Figure S1). Moreover, as expected, 2D monolayer demonstrated 1.8 fold upregulation of ALP in the presence of differentiation factors over standard media conditions by day 7, however, the expression was significantly lesser (p < 0.05) than cell-laden 3D braids (Figure S1). The expression of COL1A1, an important extracellular matrix (ECM) protein of the bone, was significantly upregulated in hFOB-laden braids by day 7 (p < 0.05) and declined by 14 days for both conditions (Figure 5); indicating temporal pattern of osteogenic differentiation.24 Negligible expression of COL1A1 (< 1 fold change) was detected in the presence of TGF-β1, both with and without differentiation factors. For 2D monolayer control (Figure S1), COL1A1 transcript levels were nominal for all conditions tested. Similarly, the expression of Runx2; master gene for bone differentiation, demonstrated drastic upregulation by day 7 in hFOB-laden braids cultured only in the presence of differentiation factors (p < 0.05) and subsequently declined by 62.2 fold by day 14; typically following the pattern of osteogenic differentiation.24 A similar pattern of expression was also observed for 2D monolayer cultured with differentiation factors (Figure S1), however, the expression levels were significantly lower than braids. In comparison, hFOB-laden braids cultured without differentiation factors depicted significantly low expression of RUNX2 at all time points (p < 0.05). Upon addition of TGF-β1, Runx2 transcript levels were found to be significantly down regulated by day 14 (p < 0.05) for both conditions as compared to groups where no TGF-β1 was added.

ACS Paragon Plus Environment

14

ACS Biomaterials Science & Engineering

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 38

The expression of SPARC, mid-stage ECM marker, in hFOB-laden braids depicted a similar expression pattern to Runx2, wherein drastic upregulation was observed by day 7 only in the presence of differentiation factors, which significantly declined by day 14 (52.4 fold decrease). Similarly, in the presence of TGF-β1, nominal expression of SPARC was observed in both experimental groups in braids (Figure 5) as well as monolayer (Figure S1) cultures. For SPP1 or osteopontin, a reverse trend could be identified compared to the above groups (Figure 5) with significant upregulation of SPP1 expression in hFOB-laden braids without differentiation factors at day 7 (3.8 fold) and day 14 (4.6 fold) compared to differentiation culture media. Moreover, the transcript levels were drastically upregulated (p < 0.05) in the presence of TGF-β1 by day 14. This pattern of differentiation was also observed for monolayer culture (Figure S1). The expression of BGLAP, indicative of later stages of osteoblast differentiation involving the deposition of calcified matrix, was also analysed. While the expression was significantly upregulated (p < 0.05) in the presence of differentiated factors by day 7, the transcript levels subsequently declined by day 14. An interesting trend that was observed in this particular gene was the drastically upregulated activity of BGLAP in the presence of TGF-β1 in both groups, however, the expression was maximal in the presence of differentiation factors by day 14 (p < 0.05). Similarly, in 2D monolayer culture, the expression of BGLAP was maximal in the case of differentiation factors (26 fold) and minimal (2 fold) in by day 7. The expression was either similar (differentiation conditions) or upregulated (standard media conditions) in the presence of TGF-β1. 3.2.4.2 Osteocytic markers - Podoplanin, an early osteocytic marker was drastically upregulated in hFOB-laden braids cultured both with (826 fold) and without differentiation

ACS Paragon Plus Environment

15

Page 17 of 38

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 Biomaterials Science & Engineering

factors (333.1 fold) by day 7, the subsequent expression differed significantly between the two groups (Figure 5). While the expression significantly declined to 129.8 fold in the presence of differentiation factors, an upregulation by 5.1 times was noticed without differentiation factors by day 14. The addition of TGF-β1 triggered the podoplanin expression by 1.5 times, albeit only in the hFOB-laden braids cultured with differentiation media. In comparison, the 2D monolayer culture demonstrated negligible expression of podoplanin by day 7 suggesting no evidence of transformation to osteocytic phenotype in monolayer condition. DMP1, an important regulator of matrix mineralization, demonstrated maximal expression in the case of hFOB-laden braids cultured in the presence of TGF-β1 with (123.6 fold) and without (98.4 fold) differentiation factors as compared to their non-TGF-β1 counterparts by day 14. hFOB-laden braids cultured in standard culture media without differentiation factors demonstrated nominal expression throughout the culture period. As DMP1 expresses in the later osteocytic stages of osteoblast to osteocyte transition, its expression was not evident in the early days of osteoblast differentiation (i.e. upto day 7) in all groups. Similar to podoplanin (Figure 5), the expression of DMP1 was negligible in 2D monolayer culture. SOST; suggesting terminal osteocytic phenotype, showed a similar trend to DMP1 (Figure 5) transcript levels. The hFOB-laden braids cultured without differentiation factors showed negligible expression throughout the culture period, however, the expression got significantly (p < 0.05) upregulated in the presence of TGF-β1. However, in the presence of differentiation factors, the expression levels of SOST were more or less comparable, indicating no significant contribution of the presence of TGF-β1 monolayer culture (Figure S1) showed negligible expression of this osteocytic marker. That fact that osteocytic expression was only evident in hFOB-laden braids and not

ACS Paragon Plus Environment

16

ACS Biomaterials Science & Engineering

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 38

monolayer culture emphasizes on the role of 3D environment and silk surface chemistry for providing essential biomechanical, topographical and chemical cues for silk matrix-mediated cell signaling. 3.2.5 Dose dependent response of TGF-β1 - A biphasic response of TGF-β1 was observed on hFOB-laden A.mylitta braids (Figure 6A), as reported earlier.25 The expression of βcatenin showed a steep increase at day 7, albeit only in TGF-β1 concentrations ranging from 2 to 6 ng/ml. By day 14, β-catenin expression declined in these groups showing significantly reduced expression (p < 0.05). However, in concentrations > 6 ng/ml, a slightly different trend was observed. A gradual increase in β-catenin expression was noticed demonstrating 22 fold and 59.7 fold upregulation at 8 ng/ml (Figure 6A) and 10 ng/ml (Figure 6B) respectively after 14 days. Based on these results, the desired concentration of 10 ng/ml was selected. Controls, consisting of hFOB-laden braids cultured in standard culture media without additional differentiation factors or TGF-β1 demonstrated similar pattern to groups containing lower dosages of TGF-β1 (< 6 ng/ml), albeit with lower expression levels (maximal 5.8 fold increase by day 7) (Figure 6C). As expected, 2D monolayer culture depicted nominal expression of β-catenin after 7 days (Figure 6C). 3.2.5.1 hFOB-laden A. mylitta braids regulate osteogenesis through Wnt/β β -catenin signaling - As reported earlier,25 we found that TGF-β1 demonstrated a dose dependent response on the expression of β-catenin (Figure 6A; Section 3.2.5), hence validating its involvement in regulating Wnt/β-catenin signaling on hFOB-laden braids. As evident in later stages of differentiation, β-catenin demonstrated maximal upregulation in TGF-β1 supplemented groups (59.7 and 100.4 fold with and without differentiation factors respectively) by day 14 (p < 0.05) which would in turn dictate the dependency of hFOBs to follow osteogenic signaling via Wnt/β-catenin signaling in the presence of TGF-β1. Although, drastic increase

ACS Paragon Plus Environment

17

Page 19 of 38

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 Biomaterials Science & Engineering

in β-catenin expression was also noted by day 7 (121.7 fold) even without TGF-β1 addition, the expression decreased by day 14, suggesting the requirement of additional factors in promoting the end-term differentiation of cells on A.mylitta braids. In contrast, negligible expression of β-catenin was observed in the 2D monolayer culture after 14 days (Figure 6C). 3.2.6 Fluorescence microscopy - In the absence of differentiation factors, cell morphology was more rounded (Figure 7A) and actin stress fibres were not very evident by day 7, as observed in SEM micrographs (Figure 4 J-L). On the contrary, hFOBs cultured in the presence of differentiation factors demonstrated extensively elongated stress fibres (Figure 7B; yellow arrows) resembling dendritic processes associated with transition to osteocytic phenotype, as also evident in SEM (Figure 4 M-O) and RT-PCR (Figure 5). Further, the synthesis of osteopontin was localized to majority of hFOBs with discrete patches of staining evident within the hFOB-laden braids in both differentiation conditions, albeit expression was relatively higher in the absence of differentiation factors by day 7 (Figure 7C) as also evident in the gene expression analysis (Figure 5; SPP1). On the contrary, the synthesis of osteonectin by day 7 was mostly localized to hFOB-laden braids cultured in the presence of differentiation factors (Figure 7 E&F) similar to gene expression data (Figure 5; SPARC). However, one limitation with immunofluorescence of silk is the inherent auto-fluorescence arising from the aromatic amino acids (tyrosine, tryptophan),26 as seen by the blue staining of silk fibres. Discussion To our knowledge, this is the first report on the development of a braided hierarchy, constituted of native (non-regenerated) non-mulberry silk fibres, as a potential scaffold for bone tissue engineering. The braids fabricated in this study mitigate the lack of adequate osteoconductivity associated with regenerated silk scaffolds while avoiding the fabrication challenges associated with most tissue engineered scaffolds, as the dissolution of this variety of silk is a difficult task to achieve.2 These native fibres of non-mulberry silk would be more osteogenic

ACS Paragon Plus Environment

18

ACS Biomaterials Science & Engineering

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 38

compared to regenerated silk scaffolds, as the protein assembly has not been disrupted during manufacturing.4 Other reported studies on A.mylitta scaffolds, such as with conventional freeze drying,2 have reconstituted dissolved silk fibres that unfortunately result in decreased apatite deposition due to the disrupted protein chains as a result of processing parameters.4,5 Moreover, the method of textile braiding is advantageous due to its ease of fabrication, customizability, cost-effectiveness. We extensively investigated the mechanics of this braided architecture of A.mylitta silk fibres, as mechanical properties can act as a potential limiting factor in context of bone engineering scaffolds, especially because bone is a load-bearing tissue. The suitability of A.mylitta as a potential scaffold for bone partially arises from its exceptional mechanical properties (stiffness of 1.41 MPa) in the range of cancellous bone (0.5−14.6 MPa).27 As reported in our previous study, the rate of degradation of A.mylitta silk is incomparably higher than the mulberry varieties (B.mori) both in vitro and in vivo.2 Moreover, the size of the pores; a critical feature of osteoconductive materials, is comparable to that of a human osteon (223 µm)19 while still retaining appropriate moduli. The tailored braid angle, which was in the range of 35-40°, also aided in sufficient compaction of the braided structure, hence adding to the mechanical properties. Taken together, this data clearly indicates that said mechanical properties of the braid used here is favorable for implantation studies. Subsequently, we examined important aspects of cell-matrix interactions between human pre-osteoblasts and the A.mylitta braids taking into consideration; cell adhesion and distribution that influences subsequent cell behavior, metabolic activity, cell morphology, gene expression analysis and protein synthesis. The SEM and AFM results on pore size and surface roughness respectively indicated topographical relevance of the A.mylitta fibres that could enhance cell adhesion, orientation and infiltration at different orders of magnitude− fibre assembly shown at the micrometer level by SEM and surface roughness as demonstrated by

ACS Paragon Plus Environment

19

Page 21 of 38

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 Biomaterials Science & Engineering

AFM in the order of nanometer scale. After analysing the cellular distribution profile on braided surface we found that osteoblasts were localised both within the pore spaces as well as also oriented parallel to the longitudinal axis of the fibres. Overall, from the cell distribution data, it is logical to speculate these two important aspects of bone formation: bone formation occurring from the infiltrated osteoblasts within pores which can regenerate bone from the scaffold’s interior, as well as cellular alignment along the fibres as they orient along the periphery of fibres forming bone from outside. However, this hypothesis warrants in vivo investigations to provide a clearer picture on the mechanism of bone formation as a function of braided geometry. Moreover, the data demonstrates relatively high cell adhesion (lesser than 2D petri dish) and orientation of seeded osteoblasts in compliance with the surface chemistry and braid morphology. However, the values depicted for cell attachment (Table III) include both initial adhesion of hFOBs on braids as well as proliferation upto 1 day, nevertheless, the fact that native silk fibroin consisting of a specific sequence of amino acids leads to enhanced cellular responses, alongwith other factors including surface roughness, rigidity, surface chemistry or all of these combined, remains true. Another interesting revelation made by a recent study was that mechanical stiffness plays a crucial role in determining the fate of osteoprogenitors.23 While softer matrices (0.58 kPa) facilitated osteocytic differentiation of seeded MC3T3-E1 progenitors on gelatin-based 3D hydrogels, stiffer matrices (1.47 kPa), on the contrary, retained the osteoblastic phenotype uptil 56 days in culture. However, it will be difficult to compare results as the material and cell line used are entirely different. The gene expression data clearly demonstrate that A.mylitta braids are innately osteoconductive and enable rapid differentiation of cultured pre-osteoblasts to mature osteocytes within 2weeks. This was also evident from the extensively elongated dentritic cellular extensions as shown by SEM and confocal imaging data; a characteristic feature of mature osteocytic

ACS Paragon Plus Environment

20

ACS Biomaterials Science & Engineering

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 38

phenotype. The Ct (copy number) values reported are normalized to expression in 2D monolayer, cultured under similar conditions. As expected, gene expression of cell-laden braids cultured in the presence of pro-osteogenic factors increased, compared to that in A.mylitta braids alone or 2D monolayer. Interestingly, transient expression of osteocytic differentiation was evident in cell-laden A.mylitta braids alone despite the absence of pro-osteogenic differentiation conditions. The observation that A.mylitta structures correlate closely with previously reported in vivo studies where 3D porous scaffolds of A.mylitta successfully regenerated critical sized defects of rat calvariae,2 an effect which might be more enhanced in the natively spun braids, demonstrates the osteogenic ability of the material in directing cellular differentiation, since no differentiation was evident in 2D monolayer cultures. Moreover, significant changes in the gene expression profile of cell-laden A.mylitta braids following supplementation with TGF-β1 further intrigued a series of questions about the role of underlying signaling mechanisms involved. Note that to be able to identify the exact signaling pathway involved, the cell behavior on braids was extensively analysed under varied experimental parameters including temperature, pro-osteogenic factors and presence or absence of nonmulberry braids (2D vs 3D), which is discussed in detail below. Wnt/β β -Catenin signaling in regulating mineralized bone differentiation on A.mylitta braids : An interesting observation was the rapid transition of hFOBs on A.mylitta braids into osteocytic phenotype within 2 weeks of culture; a phenomenon not yet reported on silk-based matrices. While the A.mylitta braids alone (without osteogenic differentiation factors) facilitated transition of hFOBs towards osteocytic morphology, this was only transient. Clearly, in the absence of osteogenic factors, while early to mid osteoblast differentiation in hFOB-laden braids (ALP, COLA1, SPP1) increased with time, the transition towards mineralized osteoblasts (osteocalcin) was highly suppressed or suboptimal. While by passing this critical stage

ACS Paragon Plus Environment

21

Page 23 of 38

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 Biomaterials Science & Engineering

of mineralization responsible for calcified bone formation in situ, the cultured hFOBs expressed drastically upregulated levels of osteocytic markers within 7-14 days of culture suggesting rapid transition of hFOBs into osteocytic cells, a phenomenon not yet reported for hFOBs even under extended culture periods.14 While early osteocyte gene continued to increase (podoplanin), we found nominal expression of Dmp1 (which participates in the process of matrix mineralization)28 at all time points studied. A possible explanation could be the reduced production of associated cytokines/growth factors involved in mineralization, which was also validated from the nominal osteocalcin expression. However, within the commercial setup, this rapid transition of pre-osteoblasts to transitory osteocytes may not serve as an ideal model of bone differentiation. Therefore, standard osteogenic reagents (Dexamethasone, Ascorbic acid, β-glycerophosphate) were added to improve the extent of osteogenesis in cellladen braids. The addition of factors convincingly upregulated the expression of osteogenic markers over A.mylitta braids alone. Another major reason for the suppressed activity of mineralized bone apatite could be the upregulated expression of Sost gene. It is known that sclerostin, secreted by mature osteocytes, binds to Lrp5 to antagonize the action of Wnt hence inhibiting Wnt/β-catenin signaling.29,30 Literature suggests that deletion of Sost gene in mice resulted in increased bone mass and strength;31 whereas, overexpression of SOST demonstrated lower bone mass.32 Similarly, loss of SOST expression in humans is often associated with bone disorders associated with high bone mass such as Van Buchem's disease33 and sclerosteosis34. Hence it was logical to believe that osteocyte secreted SOST caused the inhibition of mineralized bone tissue by blocking Wnt/Lrp signaling in osteoblasts.28,29 This mechanism resulted in lack of apatite deposition on non-mulberry silk braids. Under physiological conditions of bone differentiation, accumulation of β-catenin starts in the cytoplasm subsequently translocating into the cell nucleus wherein it interacts

ACS Paragon Plus Environment

22

ACS Biomaterials Science & Engineering

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 38

with the Tcf/Lef family of transcription factors. This interaction further regulates the activation of several important genes associated with proliferation, differentiation and apoptosis of osteogenic cells.35 Therefore to examine if this arrest of osteoblast differentiation observed at the later stages is possibly due to compromised Wnt/β-catenin signaling36,37, we supplemented the cultures with TGF-β1. It is well known that activation of Wnt/β-catenin signaling pathway results in decrease of apoptosis in osteogenic cells.35 In this context, the first indicator for the enhanced activity of Wnt/β-catenin signaling in the case of TGF-β1 supplemented braids was the increased metabolic activity found in hFOB-laden braids even when cultured at 39°C; restrictive temperature of the cell line. hFOB-laden braids cultured in TGF-β1 were found to restore the antagonizing action of SOST, thus validating the direct involvement of Wnt/β-catenin signaling in hFOBs cultured on non-mulberry silk braids. The expression of both mineralization associated genes; BGLAP and DMP1 was found to be restored suggesting natural process of bone differentiation. Moreover, the significant relation between SOST and β-catenin was established by the reciprocal levels of expression observed in hFOB-laden braids cultured without any differentiation factors. With a 8.5 fold decline in the SOST expression level, the corresponding increase in β-catenin was roughly 23 fold after 14 days in culture. However, the addition of TGF-β1 successfully upregulated the expression of βcatenin hence triggering Wnt/β-catenin signaling in regulating osteogenesis in hFOB-laden A.mylitta braids in vitro, with increased SOST expression suggesting a more stable phenotypic transition. Furthermore, it is often contemplated that a cross talk with the prostaglandin pathway activates Wnt/β-catenin pathway in osteocytes under mechanical stimulus subsequently resulting in the down regulation of Sost.35 Since no mechanical load was applied to the cellladen braids in our case, the SOST levels were constantly upregulated while simultaneously inhibiting mineralized bone differentiation. The compelling evidence indicates that supple-

ACS Paragon Plus Environment

23

Page 25 of 38

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 Biomaterials Science & Engineering

mentation of TGF-β1 increases mineral bone in hFOB-laden A.mylitta braids in vitro while simultaneously upregulating SOST to ensure the transition of pre-osteoblasts into phenotypically stable osteocytes within 2 weeks of culture period. Beyond the scope of this study, extensive research may be needed in determining how osteocytes perceive and enable signal transduction in response to such fibrous hierarchy. Moreover, mediation of the physical attributes of textile braid such as number of fibres per yarn, thickness of yarns, pore size, braid angle, rigidity as well as surface chemistry (comparison of different silk species) and their impact on cellular differentiation may provide a more comprehensive overview for controlling osteocyte differentiation for long-term sustainable cultures. Also, application of mechanical stimulus and/or bioreactor system for facilitating proper nutrient diffusion and cellular infiltration under load bearing conditions will aid in providing a biequivalent model replicating in vivo microenvironment more closely.

Fur-

thermore, the co-existence of osteoblasts and osteocytes, as observed in A.mylitta braids cultured with osteogenic factors, their relevance and cross talk in facilitating cell-cell communication for osteogenic differentiation needs to be established. As evident, targeting the Wnt/βcatenin pathway because of its critical role in controlling osteocyte- related functions i.e. regulation of bone mass strength, holds definitive potential for developing new paradigms and therapeutic applications for treating bone diseases in near future.

Conclusion Recently, much attention is being given to osteocyte-related biology as its significance in regulating the structural and functional mechanisms of bone hierarchy is being realized. Having said that, there still remain several obstacles which need to be addressed to fully untap the potential of this field of research. In conclusion, hFOB-laden A.mylitta braids display all of the characteristics of an in vitro 3D osteocytic model including the extensive den-

ACS Paragon Plus Environment

24

ACS Biomaterials Science & Engineering

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 38

dritic processes, lacunocanalicular structure, distinct genetic expression, and responses to growth factors (TGF-β1), all indicating the development of a mineralized bone-like microenvironment. These cells on A.mylitta braids alone rapidly induce the differentiation of preosteoblasts into osteocytes, albeit this transition was only transient. The additional of osteogenic differentiation factors significantly improved the expression of osteogenic-related markers marking a more discrete process of bone differentiation. However, the combinatorial effect of osteogenic factors with TGF-β1 and non-mulberry braided matrix resulted in completely establishing the process of osteogenic differentiation from pre-osteoblasts to terminal osteocytes expressing functional sclerostin regulated via Wnt/β-catenin signaling. This makes A.mylitta braided constructs an ideal model system for studying in vitro bone differentiation. Acknowledgements This study was funded by Department of Science and Technology, Science and Engineering Board,

India

(YSS/2014/000472)

and

Department

of

Biotechnology,

India

(BT/PR8038/MED/32/303/2013). References (1) (2)

(3) (4)

(5)

(6)

Melke, J.; Midha, S.; Ghosh, S.; Ito, K.; Hofmann, S. Silk fibroin as biomaterial for bone tissue engineering. Acta Biomaterialia. 2016, 31, 1-16. Sahu, N.; Baligar, P.; Midha, S.; Kundu, B.; Bhattacharjee, M.; Mukherjee, S.; Mukherjee, S.; Maushart, F.; Das, S.; Loparic, M.; et al. Nonmulberry Silk Fibroin Scaffold Shows Superior Osteoconductivity Than Mulberry Silk Fibroin in Calvarial Bone Regeneration. Adv. Healthc. Mater. 2015, 4 (11), 1709–1721 DOI: 10.1002/adhm.201500283. Midha, S.; Murab, S.; Ghosh, S. Osteogenic signaling on silk-based matrices. Biomaterials. 2016, 97, 133–153. Midha, S.; Tripathi, R.; Geng, H.; Lee, P. D.; Ghosh, S. Elucidation of differential mineralisation on native and regenerated silk matrices. Mater. Sci. Eng. C 2016, 68, 663–674 DOI: 10.1016/j.msec.2016.06.041. Koebley, S. R.; Thorpe, D.; Pang, P.; Chrisochoides, P.; Greving, I.; Vollrath, F.; Schniepp, H. C. Silk reconstitution disrupts fibroin self-assembly. Biomacromolecules 2015, 16 (9), 2796–2804 DOI: 10.1021/acs.biomac.5b00732. Marx, V. Models: stretching the skills of cell lines and mice. Nat. Methods 2014, 11 (6), 617–620 DOI: 10.1038/nmeth.2966.

ACS Paragon Plus Environment

25

Page 27 of 38

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 Biomaterials Science & Engineering

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18) (19) (20)

(21)

(22)

Jung, S.R.; Song, N.J.; Yang, D. K.; Cho, Y.J.; Kim, B.J.; Hong, J.W.; Yun, U. J.; Jo, D.G.; Lee, Y. M.; Choi, S. Y.; et al. Silk proteins stimulate osteoblast differentiation by suppressing the Notch signaling pathway in mesenchymal stem cells. Nutr. Res. 2013, 33 (2), 162–170 DOI: 10.1016/j.nutres.2012.11.006. Wu, M.; Chen, G.; Li, Y.P. TGF-β and BMP signaling in osteoblast, skeletal development, and bone formation, homeostasis and disease. Bone Res. 2016, 4, 16009 DOI: 10.1038/boneres.2016.9. Tong, S.; Xu, D.P.; Liu, Z.M.; Du, Y.; Wang, X.K. Synthesis of and in vitro and in vivo evaluation of a novel TGF-β1-SF-CS three-dimensional scaffold for bone tissue engineering. Int. J. Mol. Med. 2016 DOI: 10.3892/ijmm.2016.2651. Zellin, G.; Beck, S.; Hardwick, R.; Linde, A. Opposite effects of recombinant human transforming growth factor-beta 1 on bone regeneration in vivo: effects of exclusion of periosteal cells by microporous membrane. Bone 1998, 22 (6), 613–620. Hulsart-Billström, G.; Dawson, J. I.; Hofmann, S.; Müller, R.; Stoddart, M. J.; Alini, M.; Redl, H.; El Haj, A.; Brown, R.; Salih, V.; et al. A surprisingly poor correlation between in vitro and in vivo testing of biomaterials for bone regeneration: Results of a multicentre analysis. Eur. Cells Mater. 2016, 31, 312–322. Yen, M.L.; Chien, C.C.; Chiu, I.M.; Huang, H.I.; Chen, Y.C.; Hu, H.I.; Yen, B. L. Multilineage differentiation and characterization of the human fetal osteoblastic 1.19 cell line: a possible in vitro model of human mesenchymal progenitors. Stem Cells 2007, 25 (1), 125–131 DOI: 10.1634/stemcells.2006-0295. Harris, S. a; Enger, R. J.; Riggs, B. L.; Spelsberg, T. C. Development and characterization of a conditionally immortalized human fetal osteoblastic cell line. J. Bone Miner. Res. 1995, 10 (2), 178–186 DOI: 10.1002/jbmr.5650100203. Keogh, M. B.; O’Brien, F. J.; Daly, J. S. A novel collagen scaffold supports human osteogenesis - Applications for bone tissue engineering. Cell Tissue Res. 2010, 340 (1), 169–177 DOI: 10.1007/s00441-010-0939-y. Fielding, G. A.; Bandyopadhyay, A.; Bose, S. Effects of silica and zinc oxide doping on mechanical and biological properties of 3D printed tricalcium phosphate tissue engineering scaffolds. Dent. Mater. 2012, 28 (2), 113–122 DOI: 10.1016/j.dental.2011.09.010. Tarafder, S.; Balla, V. K.; Davies, N. M.; Bandyopadhyay, A.; Bose, S. Microwavesintered 3D printed tricalcium phosphate scaffolds for bone tissue engineering. J. Tissue Eng. Regen. Med. 2013, 7 (8), 631–641 DOI: 10.1002/term.555. Chameettachal, S.; Midha, S.; Ghosh, S. Regulation of chondrogenesis and hypertrophy in silk fibroin-gelatin based 3D bioprinted constructs. ACS Biomater. Sci. Eng. 2016, acsbiomaterials.6b00152 DOI: 10.1021/acsbiomaterials.6b00152. Petrulis, D.; Petrulyt, S. Hypothetical Models of Yarn Cross-section : Comparison of the Shapes and Packing Peculiarities. 2003, 9 (4). Holmes, R. E. Bone regeneration within a coralline hydroxyapatite implant. Plastic and reconstructive surgery 1979, 626–633. Kim, H. J.; Kim, U.J.; Vunjak-Novakovic, G.; Min, B.H.; Kaplan, D. L. Influence of macroporous protein scaffolds on bone tissue engineering from bone marrow stem cells. Biomaterials 2005, 26 (21), 4442–4452 DOI: 10.1016/j.biomaterials.2004.11.013. Ghasemi-Mobarakeh, L. Structural properties of scaffolds: Crucial parameters towards stem cells differentiation. World J. Stem Cells 2015, 7 (4), 728 DOI: 10.4252/wjsc.v7.i4.728. Liu, X.; Lim, J. Y.; Donahue, H. J.; Dhurjati, R.; Mastro, A. M.; Vogler, E. A. Influence of substratum surface chemistry/energy and topography on the human fetal oste-

ACS Paragon Plus Environment

26

ACS Biomaterials Science & Engineering

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

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35) (36)

Page 28 of 38

oblastic cell line hFOB 1.19: Phenotypic and genotypic responses observed in vitro. Biomaterials 2007, 28 (31), 4535–4550 DOI: 10.1016/j.biomaterials.2007.06.016. McGarrigle, M. J.; Mullen, C. A.; Haugh, M. G.; Voisin, M. C.; McNamara, L. M. Osteocyte differentiation and the formation of an interconnected cellular network in vitro. Eur. Cells Mater. 2016, 31, 323–340 DOI: 10.22203/eCM.v031a21. Stein, G. S.; Lian, J. B. Molecular mechanisms mediating proliferation/differentiation interrelationships during progressive development of the osteoblast phenotype. Endocrine Reviews 1993, 424–442. Zhang, H.; Ahmad, M.; Gronowicz, G. Effects of transforming growth factor-beta 1 (TGF-β1) on in vitro mineralization of human osteoblasts on implant materials. Biomaterials 2003, 24 (12), 2013–2020 DOI: 10.1016/S0142-9612(02)00616-6. Abbott, R. D.; Kimmerling, E. P.; Cairns, D. M.; Kaplan, D. L. Silk as a Biomaterial to Support Long-Term Three-Dimensional Tissue Cultures. ACS Appl. Mater. Interfaces 2016, acsami.5b12114 DOI: 10.1021/acsami.5b12114. Hung, B. P.; Naved, B. A.; Nyberg, E. L.; Dias, M.; Holmes, C. A.; Elisseeff, J. H.; Dorafshar, A. H.; Grayson, W. L. Three-Dimensional Printing of Bone Extracellular Matrix for Craniofacial Regeneration. ACS Biomater. Sci. Eng. 2016, 2 (10), 1806– 1816 DOI: 10.1021/acsbiomaterials.6b00101. Thompson, W. R.; Uzer, G.; Brobst, K. E.; Xie, Z.; Sen, B.; Yen, S. S.; Styner, M.; Rubin, J. Osteocyte specific responses to soluble and mechanical stimuli in a stem cell derived culture model. Sci. Rep. 2015, 5, 11049 DOI: 10.1038/srep11049. Tu, X.; Rhee, Y.; Condon, K. W.; Bivi, N.; Allen, M. R.; Dwyer, D.; Stolina, M.; Turner, C. H.; Robling, A. G.; Plotkin, L. I.; et al. Sost downregulation and local Wnt signaling are required for the osteogenic response to mechanical loading. Bone 2012, 50 (1), 209–217 DOI: 10.1016/j.bone.2011.10.025. McClung, M. R.; Grauer, A.; Boonen, S.; Bolognese, M. A.; Brown, J. P.; Diez-Perez, A.; Langdahl, B. L.; Reginster, J.-Y.; Zanchetta, J. R.; Wasserman, S. M.; et al. Romosozumab in postmenopausal women with low bone mineral density. N. Engl. J. Med. 2014, 370 (5), 412–420 DOI: 10.1056/NEJMoa1305224. Li, X.; Ominsky, M. S.; Niu, Q.T.; Sun, N.; Daugherty, B.; D’Agostin, D.; Kurahara, C.; Gao, Y.; Cao, J.; Gong, J.; et al. Targeted deletion of the sclerostin gene in mice results in increased bone formation and bone strength. J. Bone Miner. Res. 2008, 23 (6), 860–869 DOI: 10.1359/jbmr.080216. Rhee, Y.; Allen, M. R.; Condon, K.; Lezcano, V.; Ronda, A. C.; Galli, C.; Olivos, N.; Passeri, G.; O’Brien, C. A.; Bivi, N.; et al. PTH receptor signaling in osteocytes governs periosteal bone formation and intracortical remodeling. J. Bone Miner. Res. 2011, 26 (5), 1035–1046 DOI: 10.1002/jbmr.304. Balemans, W.; Ebeling, M.; Patel, N.; Van Hul, E.; Olson, P.; Dioszegi, M.; Lacza, C.; Wuyts, W.; Van Den Ende, J.; Willems, P.; et al. Increased bone density in sclerosteosis is due to the deficiency of a novel secreted protein (SOST). Human molecular genetics 2001, 537–543. Brunkow, M. E.; Gardner, J. C.; Van Ness, J.; Paeper, B. W.; Kovacevich, B. R.; Proll, S.; Skonier, J. E.; Zhao, L.; Sabo, P. J.; Fu, Y.; et al. Bone dysplasia sclerosteosis results from loss of the SOST gene product, a novel cystine knot-containing protein. Am. J. Hum. Genet. 2001, 68 (3), 577–589 DOI: 10.1086/318811. Bonewald, L. F.; Johnson, M. L. Osteocytes, mechanosensing and Wnt signaling. Bone 2008, 606–615. Holmen, S. L.; Zylstra, C. R.; Mukherjee, A.; Sigler, R. E.; Faugere, M.-C.; Bouxsein, M. L.; Deng, L.; Clemens, T. L.; Williams, B. O. Essential role of beta-catenin in

ACS Paragon Plus Environment

27

Page 29 of 38

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 Biomaterials Science & Engineering

(37)

postnatal bone acquisition. J. Biol. Chem. 2005, 280 (22), 21162–21168 DOI: 10.1074/jbc.M501900200. Hu, H.; Hilton, M. J.; Tu, X.; Yu, K.; Ornitz, D. M.; Long, F. Sequential roles of Hedgehog and Wnt signaling in osteoblast development. Development 2005, 132 (1), 49–60 DOI: 10.1242/dev.01564.

ACS Paragon Plus Environment

28

ACS Biomaterials Science & Engineering

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 Paragon Plus Environment

Page 30 of 38

Page 31 of 38

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 Biomaterials Science & Engineering

Figure 1: Physical characterization of A.mylitta braid; (A) Schematic explaining the structur-al braid hierarchy and nomenclature established; (B-E) SEM micrographs of braid demon-strating a part of the braided fabric (B,C), two plied yarn (D) and silk fibres (E). Figure 1 219x166mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering

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

Figure 2: Surface roughness of silk yarn (n=3) by atomic force microscopy displaying height sensor, phase and 3D micrograph. Figure 2 180x189mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 32 of 38

Page 33 of 38

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 Biomaterials Science & Engineering

Figure 3: MTT of hFOB-laden A.mylitta braids measured at different temperatures. ‘*’ rep-resents statistically significant data where p < 0.05, n=3/group. Figure 3 190x142mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering

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

Figure 4: SEM micrographs of hFOB-laden A.mylitta braids imaaged at (A-C) Day 1 and (D-O) Day 7 post cell seeding. Morphological differences in cellular morphology and extracellu-lar matrix components were visualized at day 7 in cell-laden braids cultured at; 33.5°C with (G-I) and without (D-F) differentiation factors; 39°C with (M-O) and without (J-L) differen-tiation factors. Abbreviations: w/o factors (without osteogenic differentiation factors); with factors (osteogenic differentiation factors). Scale Bar for inset (E) = 10 µm. Figure 4 254x309mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 34 of 38

Page 35 of 38

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 Biomaterials Science & Engineering

Figure 5: Gene expression analysis of hFOB-laden A.mylitta braids (n=3/group) cultured at 39°C in different experimental conditions upto day 14. Abbreviations: w/o (without osteogenic factors/standard media conditions); w/o + TGF (without osteogenic factors, with 10 ng/ml TGF-β1); with (with osteogenic factors); with + TGF (with osteogenic factors, with 10 ng/ml TGF-β1). 254x196mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering

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

Figure 6: Gene expression analysis of β-catenin (key regulator of Wnt signaling) demonstrat-ing; (A) Dose dependent response of TGF-β1 on cell-laden braids; (B,C) β-catenin expres-sion monitored at the selected concentration of TGF-β1 (10 ng/ml) with respect to braids (B) and 2D monolayer (C) cultures. The reactions were set up in triplicate. Figure 6 204x154mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 36 of 38

Page 37 of 38

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 Biomaterials Science & Engineering

Figure 7: Confocal micrographs of hFOB-laden braids after 7 days; (A) actin (red); and (B) osteopontin (red). Blue color represents nuclear staining. The silk fibres visible in blue are due to autofluorescence. Scale Bars = 50 µm. Figure 7 254x309mm (300 x 300 DPI)

ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering

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

Table of Contents graphic: Till date, the development of physiologically relevant and phenotypically stable 3D in vitro model of differentiated bone tissue remains elusive. The present work aims to develop a 3D in vitro model of bone differentiation on non-mulberry Antheraea mylitta textile-based silk braids and explore the underlying signaling mechanism responsible for the transition from pre-osteoblast to terminal osteocytic phenotype. We conclude that the unique combination of substrate properties (braid geometry, rigidity, surface roughness, surface chemistry) with osteogenic differentiation conditions supported extended viability of cultured human osteoblasts further inducing rapid terminal osteocytic differentiation within 2 weeks. Moreover, TGF-β1 regulated Wnt/β-catenin signaling pathway played a pivotal role in stabilizing differentiation of mineralized osteoblasts towards mature osteocytic pheotype. The knowledge and application of this understanding to tissue engineering of bone using silk biomaterials can help develop more relevant in vitro assay models for therapeutic applications in screening the efficacy of drugs. Tables of Contents Graphic 254x127mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 38 of 38