Bio-inspired in situ crosslinking and mineralization of electrospun collagen scaffolds for bone tissue engineering

Accepted Manuscript Bio-Inspired in situ crosslinking and mineralization of electrospun collagen scaffolds for bone tissue engineering Chetna Dhand, Seow Theng Ong, Neeraj Dwivedi, Silvia M. Diaz, Jayarama R. Venugopal, Balchandar Navaneethan, Mobashar H.U.T. Fazil, Shouping Liu, Vera Seitz, Erich Wintermantel, Roger W. Beuerman, Prof. Seeram Ramakrishna, Asst. Prof. Navin K. Verma, Asst. Prof. Rajamani Lakshminarayanan PII:

S0142-9612(16)30338-6

DOI:

10.1016/j.biomaterials.2016.07.007

Reference:

JBMT 17611

To appear in:

Biomaterials

Received Date: 24 June 2016 Accepted Date: 4 July 2016

Please cite this article as: Dhand C, Ong ST, Dwivedi N, Diaz SM, Venugopal JR, Navaneethan B, Fazil MHUT, Liu S, Seitz V, Wintermantel E, Beuerman RW, Ramakrishna S, Verma NK, Lakshminarayanan R, Bio-Inspired in situ crosslinking and mineralization of electrospun collagen scaffolds for bone tissue engineering, Biomaterials (2016), doi: 10.1016/j.biomaterials.2016.07.007. 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.

ACCEPTED MANUSCRIPT

Bio-Inspired in situ Crosslinking and Mineralization of Electrospun Collagen Scaffolds for Bone Tissue Engineering Chetna Dhand,1 Seow Theng Ong,2 Neeraj Dwivedi,3 Silvia M. Diaz,4,5 Jayarama R. Venugopal,4 Balchandar Navaneethan,4 Mobashar H. U. T. Fazil,2 Shouping Liu,1,6 Vera Seitz,5 Erich Wintermantel,5 Roger W. Beuerman,1,6 Seeram Ramakrishna,*,4,7 Navin K. Verma,*,2 and Rajamani Lakshminarayanan*,1,6 1

Anti-Infectives Research Group, Singapore Eye Research Institute, The Academia, 20

College Road, Discovery Tower, Singapore 169856 2

Lee Kong Chian School of Medicine, Nanyang Technological University, Experimental

Medicine Building, Singapore 636921 3

Department of Electrical and Computer Engineering, National University of Singapore, 3

Engineering Drive 3, Singapore 117583 4

Center for Nanofibers and Nanotechnology, Department of Mechanical Engineering,

National University of Singapore, Singapore 119260. 5

Institute of Medical and Polymer Engineering, Technische Universität München,

Boltzmannstrasse 15, 85748 Garching, Germany 6

Ophthalmology and Visual Sciences Academic Clinical Program, Duke-NUS Graduate

Medical School, Singapore 169857 7

Guangdong-Hongkong-Macau Institute of CNS Regeneration (GHMICR), Jinan University,

Guangzhou 510632, China

*Correspondence should be addressed to Asst. Prof. Rajamani Lakshminarayanan ([email protected]), Asst. Prof. Navin Kumar Verma ([email protected]) and Prof. Seeram Ramakrishna ([email protected]).

1

ACCEPTED MANUSCRIPT ABSTRACT Bone disorders are the most common cause of severe long term pain and physical disability, and affect millions of people around the world. In the present study, we report bio-inspired preparation of bone-like composite structures by electrospinning of collagen containing catecholamines and Ca2+. The presence of divalent cation induces simultaneous partial oxidative polymerization of catecholamines and crosslinking of collagen nanofibers, thus producing mats that are mechanically robust and confer photoluminescence properties. Subsequent mineralization of the mats by ammonium carbonate leads to complete oxidative polymerization of catecholamines and precipitation of amorphous CaCO3. The collagen composite scaffolds display outstanding mechanical properties with Young’s modulus approaching the limits of cancellous bone. Biological studies demonstrate that human fetal osteoblasts seeded on to the composite scaffolds display enhanced cell adhesion, penetration, proliferation, differentiation and osteogenic expression of osteocalcin, osteopontin and bone matrix protein when compared to pristine collagen or tissue culture plates. Among the two catecholamines,

mats

containing

norepinephrine

displayed

superior

mechanical,

photoluminescence and biological properties than mats loaded with dopamine. These smart multifunctional scaffolds could potentially be utilized to repair and regenerate bone defects and injuries. Keywords: Collagen, catecholamine, crosslinking, bone, electrospinning, mineralization

2

ACCEPTED MANUSCRIPT 1. INTRODUCTION Bone fracture is one of the most critical public health issues and defective bone reconstruction is considered as a leading cause of disability and morbidity.1,2 In the United States alone, more than two million people suffer from osteoporosis-related fractures, trauma, congenital bone malformations, skeletal diseases and tumor resections and about 800,000 bone graft procedures have been carried out each year with an estimated healthcare cost of US$ 17 billion.3,4 Clinical procedures such as autografts and allografts are expensive but do not guarantee positive results. In addition, there is a shortage of adequate supply for treating patients with major bone defects. Furthermore, secondary complications such as infections, graft rejection, lack of biocompatibility, and limited durability of existing materials used for bone reconstruction pose considerable risks for the patients. Bone tissue engineering seeks to develop biological substitutes that integrate three important characteristics of natural bone: osteoconduction, osteoinductivity and osteogenesis. Several key attributes, including, topographical and biochemical cues as well as surface and bulk properties of the scaffolds are taken into consideration while designing new osteoconductive scaffolds.5,6 Among the various strategies available for the fabrication of bioactive scaffolds, electrospinning is an attractive method owing to its simplicity, versatility and ability to produce wide variety of polymeric scaffolds that can mimic the features of extracellular matrix (ECM).7-10 The presence of inorganic bioactive minerals with polymers promotes bone cell attachment, proliferation, differentiation and expression of osteogenic proteins.11-16 The composite nanofibrous structures can be prepared by co-electrospinning which combines the use of natural or synthetic polymers or their blends with inorganic particles such as micro or nanoparticles of CaCO3, -tricalcium phosphate and hydroxyapatite and bioactive glasses.1117

Subsequent crosslinking of the electrospun mats generated biomimetic composite

structures with osteoconductive properties.18,19 Alternatively, surface mineralization of nanofibers or electrospraying of the minerals during electrospinning have been attempted to promote cell proliferation and differentiation.20-23 However, these approaches are time consuming as multiple steps are involved in the preparation processes and the electrospun mats often contained discontinuous nanofibers since the dope solutions encompass significantly high amount of inorganics, which facilitate the formation of beaded fibers. In addition, the methodology increased the roughness of fibre surface and there was a moderate increase in the mechanical properties of the scaffolds relative to the polymer alone. 3

ACCEPTED MANUSCRIPT It has been shown that substantial changes in the biochemical and biomechanical properties take place during the early development of vertebrate bone.24,25 The increase in numbers of intra- and inter-fibrillar crosslinking of collagen network is followed by mineralization during the maturation, thus achieving maximum stiffness and toughness. 25 By mimicking this process, we present here a versatile and easy-to-perform method enabling the formation of in situ crosslinked collagen-CaCO3 composite scaffolds which confer outstanding mechanical, photoluminescent and osteoconductive properties. In the current study, we prepared electrospun (ES) collagen (Coll) mats doped with catecholamines and CaCl2 followed by exposure to ammonium carbonate, (NH4)2CO3. The presence of divalent cations during electrospinning triggered oxidative polymerization of catecholamines, thus forming collagen network with soldered junctions. Subsequent exposure to (NH4)2CO3 further promoted oxidative polymerization of catecholamines and in situ precipitation of CaCO3. We used two different catecholamines, dopamine (DA) and norepinephrine (NE), since both can form material-independent coating under alkaline conditions.26-28 We then characterized the surface, bulk and photoluminescent properties and analyzed the biological effect of composite structures in terms of cellular adhesion, proliferation and differentiation of human fetal osteoblasts (hFob). In summary, we are reporting here a new bio-inspired in situ chemical cross-linking and mineralization strategy to design potential osteoconductive scaffolds for bone tissue engineering. The overall strategy and summary of the work is illustrated in Figure 1.

Mineralization V

Collagen + Catecholamines + CaCl2

+ + + + +

High Voltage

(NH4)2CO3 diffusion

Wettability

Electrochemical oxidation Photoluminescent

Osteoblasts differentiation

Figure 1. Overall summary of the present work. Electrospinning of a dope solution containing collagen (8% w/v), dopamine (10% w/w of collagen) and 20 mM CaCl2 in 90% HFIP triggered the formation of polydopamine supported by the appearance of brown 4

ACCEPTED MANUSCRIPT coloration in the mats. Subsequent exposure of the mats to (NH4)2CO3 vapors resulted in intensification of brown coloration and precipitation of CaCO3. For a comparison the photograph of mat electrospun without CaCl2 is also shown in the lower bottom. The nanofibrous composite mats displayed excellent mechanical properties, blue and green fluorescence, surface wettability and osteoblast cell proliferation and differentiation. 2. MATERIALS AND METHODS 2.1. Cells and Reagents: Atelocollagen powder (Collagen Type I, Product No. CLP-01) from bovine dermis was a product of Koken (Tokyo, Japan) and purchased from Unison Collaborative Pte Ltd. Dopamine hydrochloride (DA), norepinephrine hydrochloride (NE), 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), Dulbecco’s Modified Eagle’s Medium (DMEM), nutrient

mixture

F-12

(HAM),

antibiotics,

anti-goat

TRITC

antibody,

Hoechst,

glutaraldehyde, Alizarin Red-S, p-nitrophenyl phosphate, cetylpyridinium chloride (CPC), hexamethyl-disilazane (HMDS), dimethyl sulphoxide (DMSO, 99+%), triton X-100, calcium chloride (CaCl2), ammonium carbonate and sodium chloride (NaCl) were obtained from Sigma-Aldrich (Singapore). Anti-mouse Alexa Fluor 633 and Alexa Fluor 647 Phalloidin were purchased from Molecular Probe®. Anti-GAPDH antibody was obtained from Merck Millipore (S) Pte Ltd. Cytiva Cell Health Reagents were procured from GE Healthcare Life Sciences. Human fetal osteoblast cells (hFob) were obtained from the American Type Culture Collection (ATCC, Manassas, VA). For cell culture, fetal bovine serum (FBS) and trypsinEDTA were purchased from Gibco™ (ThermoFisher Scientific). CellTracker™ Green CMFDA (5-chloromethylfluorescein diacetate) and CellTiter 96® AQueous One solution were purchased from Promega (Singapore). All the above chemicals were of analytical grade and were used without further purification. 2.2. Electrospinning of Collagen and Composite Nanofibers. Pristine collagen mats were prepared from 8% (w/v) dope solution in hexafluoro isopropanol (HFIP) and the solution was transferred to a polypropylene plastic syringe with 27G stainless steel blunted needle. The solution was extruded at an applied voltage of 13 kV from a high voltage power supply (Gamma High Voltage Research, Inc., FL, USA) and the distance between needle and collector (a flattened aluminum foil) was set at 17 cm at a feed rate of 1 ml/h (KD 100 Scientific Inc., MA, USA). Identical conditions were used for the preparation of mats containing 10% (w/w) catecholamines. However, for the preparation of mats containing catecholamines and CaCl2, collagen was dissolved in 90% HFIP and 10% H2O containing CaCl2 (final concentration of 20 mM Ca2+). The voltage/distance of 17 kV/13 cm and a feed rate of 0.8 mL/h produced bead-free nanofibers. All the electrospinning experiments were 5

ACCEPTED MANUSCRIPT performed at room temperature at around 25% humidity and the collected electrospun mats were vacuum dried for 48 h to remove any residual HFIP and then stored in dry cabinets for ammonium carbonate exposure and characterization. The catecholamine-loaded mats with or without Ca2+ were placed in a sealed desiccator containing ~5 g of (NH4)2CO3 powder for 24 h to induce oxidative polymerization and precipitation of CaCO3. For SEM, photoluminescence and cell culture studies, the fibers were collected on microscopy cover slips (15 mm) and on gold-coated copper grids for TEM. For simplicity, the mats are labelled as follows: pristine collagen mats – ES_Coll; As-spun collagen mats with DA or NE Coll_DA or Coll_NE; Collagen mats after (NH4)2CO3 exposure - Coll_pDA and Coll_pNE; As-spun collagen mats containing DA or NE and 20 mM Ca2+ - Coll_DA_Ca or Coll_NE_Ca; Collagen mats containing DA or NE and 20 mM Ca2+ after (NH4)2CO3 exposure - Coll_pDA_Ca or Coll_pNE_Ca. 2.3. Morphological Characterization by Scanning and Transmission Electron Microscopies (SEM/TEM). Morphological analysis of collagen mats were investigated by field emission scanning electron microscopy (FE-SEM) to infer i) the influence of incorporating catecholamines within collagen scaffold, ii) the effect of integrating Ca2+ions in the catecholamine-loaded mats, iii) the influence of crosslinking treatment on the morphology of the electrospun collagen nanofibers, and iv) to study the fracture morphology of the various scaffolds after uniaxial tensile testing. SEM studies were also performed to check the cell adhesion and cellular morphology seeded onto the various scaffolds. SEM analysis was performed on a FE-SEM (FEI-QUANTA 200F, The Netherlands) at an accelerating voltage of 15 KV after sputter coating the samples with Platinum (JEOL JSC-1200 fine coater, Japan). Image Analysis Software (Image J, National Institute of Health, USA) was used for estimating the average fibre diameter for various scaffolds. For calculating the average diameter of each sample, ~ 100 nanofibers from their respective SEM images (3-4 micrographs focusing different areas) were randomly selected and used for measuring their diameter. Selected area electron diffraction patterns (SAED) and TEM studies were performed using JEOL JEM-3010 instrument. 2.4. X-ray Photoelectron Spectroscopy. XPS studies were carried out using Kratos AXIS UltraDLD (Kratos Analytical Ltd) in ultrahigh vacuum (UHV) conditions of ~10-9 Torr by employing a monochromatic Al-Kα X-ray source (1486.71 eV). The general scan and different high resolution spectra were recorded for in-depth analysis of various chemical states of fabricated samples. During analysis, the high resolution spectra were deconvoluted

6

ACCEPTED MANUSCRIPT using various Gaussian-Lorentzian components with the background subtracted in Shirley mode. 2.5. Determination of Mechanical Properties. A tabletop tensile tester (Instron 5345, USA) using a load cell of 10 N capacities at ambient conditions was used to carry out the tensile testing of the electrospun fibers in accordance with the ASTMD882-02 protocol. All the mechanical properties (Tensile strength, Failure Strain, Young’s Modulus and Work of Failure) were calculated based on the generated stress–strain curves for each collagen mat. The mats were cut into rectangular strips of 1 cm  3 cm and the thickness of each sample was measured using a micrometre calliper. Finally, the samples were mounted vertically on the gripping unit of the tensile tester at a cross-head speed of 5 mm min−1. The average results are reported from 2-4 independent measurements. 2.6. Determination of Water Contact Angle (WCA). Surface wettability of the ES mats was determined by dynamic water contact angle measurements on a VCA Optima Surface Analysis system (AST products, MA, USA). A 1 μl drop of distilled water was placed on the fibre mats and photographed continuously at every 10 s for 60 seconds. The linear part of the non-linear decrease in contact angle – time curve was extrapolated to 0 time to determinestatic values. The reported values were determined from two independent triplicate experiments. 2.7. Photoluminescence Properties of Nanofibrous Scaffolds. Fluorescence microscopy was performed using a confocal microscope (Zeiss LSM 800 Airyscan, Carl Zeiss Microimaging Inc., NY, USA) equipped with 405, 488, 561 and 640 nm excitation lasers. A Plan-Apochromat 63×/1.4 Oil DIC M27 objective lens (Carl Zeiss) was used and the confocal pinhole was set to 1 Airy unit for the green (488 nm) channel and other channels were adjusted to the same optical slice thickness. Images from different sample groups were acquired using identical laser percentage and digital gain for each laser channel. Zen 2.1 lite imaging software (Carl Zeiss) was used for the figure preparation. 2.8. Human Fetal Osteoblasts (hFob) cell culture. The hFob cells were cultured in DMEM/F12 medium (1:1) supplemented with 10% FBS and cocktail antibiotics in 75 cm2 cell-culture flasks. The hFob cells were incubated at 37C in humidified CO2 incubator for 1 week and fed with fresh medium every 3 days. Cells were harvested after 3rd passage using trypsin-EDTA treatment and replated after cell counting with trypan blue using haemocytometer. For cell seeding, the nanofibrous scaffolds were collected on 15 mm coverslips and sterilized under UV light for 1 h. The scaffolds were then placed in 24-well 7

ACCEPTED MANUSCRIPT plates with stainless steel rings to prevent the lifting up of the scaffolds. The scaffolds were then washed with 10 mM PBS (pH 7) thrice for 15 min to remove the residual solvent and finally soaked in complete media overnight. The hFob cells were seeded at a density of 1 × 104 cells well-1 on COLL_DA_Ca, COLL_pDA_Ca, COLL_NE_Ca and COLL_pNE_Ca scaffolds. ES_Coll and tissue culture plate (TCP) served as positive controls. 2.9. Cell Adhesion, Proliferation and Differentiation Assays. CMFDA is a cell penetrating dye and readily cleaved by the intracellular esterases present in live cells, thus producing green-fluorescent calcein. After 3, 6 or 9 days of cell growth, the complete medium was removed from the 24-well plates and the cells were fed with DMEM medium. The scaffolds were then incubated with 20 l CMFDA dye (final concentration 25 M in medium) for 2 h at 37C. Thereafter, the CMFDA-medium was removed and 1 ml complete medium was added to the cells and incubated overnight. Before imaging by confocal microscopy, cells were treated with Cytiva Cell Health Reagents for 1 h to determine cell count, nuclear morphology and cell viability. Confocal images with z-stacks were acquired with 405, 488 and 561 nm lasers excitation using Zeiss LSM800 Airyscan a Plan-Apochromat ×40/1.3 oil immersion objective lens. To quantify cell viability, images were acquired by an automated microscope IN Cell Analyzer 2200 (GE Healthcare Life Sciences), 9 randomly selected fields/samples using a ×10 objective. Quantitative live/dead cell analysis of the acquired images was performed with IN Cell Investigator software (GE Healthcare Life Sciences). To visualize the morphologies of cells growing on various scaffolds, cells were fixed with 4% (v/v) formaldehyde at day 3, 6 or 9 p.s. After fixation, cells were stained with Alexa Fluor 647-Phalloidin (Molecular Probes®) and Hoechst. Confocal images with z-stacks were acquired with 405, 488 and 640 nm lasers excitation using Zeiss LSM800 Airyscan a PlanApochromat ×40/1.3 oil immersion objective lens. Confocal images were prepared using Zen 2.1 lite imaging software (Carl Zeiss). The hFob cell proliferation on ES scaffolds was assessed by MTS (3-(4,5dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H tetrazolium, inner salt), CellTiter 96® AQueous One Assay. The absorbance readout from MTS assay was converted into cell number using a calibration curve. The cell proliferation ratio was obtained by dividing cell number at a specific time point by cell number at day 0 and reported. This assay is based on the fact that the metabolically active cells react with MTS tetrazolium salt to produce a soluble formazan dye with absorbance at 490 nm. For performing this assay, cellular constructs were first rinsed with PBS and then incubated with 20% MTS reagent in

8

ACCEPTED MANUSCRIPT serum free medium for 3 h. Thereafter, the 100 µl aliquots were transferred into 96 well plates and the absorbance readings were recorded using spectrophotometric plate reader (FLUOstar OPTIMA, BMG Lab Technologies, Germany) at 490 nm. Alkaline phosphatase (ALP) is an important enzyme responsible for the mineral nucleation by supplying free phosphate ions through lysis of organic phosphates and its expression is associated with cell differentiation. ALP activity of hFob cells seeded on various scaffolds was estimated using an alkaline phosphate yellow liquid substitute system for enzyme-linked immunosorbent assay (ELISA) (Sigma Life Science, USA). In this assay, ALP catalyses the hydrolysis of colourless p-nitro phenyl phosphate (PNPP) to a yellow product p-nitrophenol and phosphate. At 3, 6, 9, 11 and 14 days p.s., the medium was removed from the 24-well plates and the scaffolds were washed thrice with PBS. The scaffolds were then incubated with 400 l of PNPP solution for 30 min. The reaction was dragged to completion by adding 200 l of 2 M NaOH solution. The resultant yellow coloured solution was then pipetted out into the 96-wells plates and the absorbance for different scaffolds was read at 405 nm in micro-plate reader. To present the ALP activity of the cells at different time points, the ALP activity was normalized with cell number as a marker for bone formation. 2.10. Alizarin Red-S (ARS) Staining Assay. ARS staining was used to qualitatively and quantitatively detect the extent of mineralization on various scaffolds. ARS is a dye that selectively binds to the calcium salts and used for the calcium mineral histochemistry. The nanofibrous scaffolds with the hFob cells were first washed thrice using PBS and then the cells were fixed by treating them to 70% ethanol for 1 h. The cellular constructs were then washed thrice with DI water followed by staining with ARS (40 mM) for 20 min at room temperature. The scaffolds were then washed with DI water several times and visualized under optical microscope. For quantitative assessment, the stain was eluted with 10% cetylpyridinium chloride for 60 min. The absorbance of solution was recorded at 540 nm on Tecan microplate reader.

2.11. Cell Morphology Analysis by FE-SEM. Cellular morphology of in vitro cultured hFob was analysed at 11 and 14 days p.s., by FE-SEM (FEI-QUANTA 200F). Cell-seeded scaffolds were washed with PBS to remove the non-adherent cells and fixed by using 3% glutaraldehyde at room temperature. The cell scaffolds were then dehydrated using a series of graded alcohol solutions and finally dried into HMDS overnight. Dried cellular constructs 9

ACCEPTED MANUSCRIPT were then sputter-coated with platinum and observed under FE-SEM at an accelerating voltage of 10 KV. 2.12. Immunofluorescent Staining for Osteocalcin (OC), Osteopontin (OPN) and Bone Morphogenetic Protein (BMP) and Western Immunoblot Analysis. The hFob cells cultured on various scaffolds and TCP were processed for protein staining 11 days p.s. Cells were fixed in 4% (v/v) formaldehyde for 10 min, followed by permeabilization using 0.3% Triton X-100 for 5 min. Cells were then labeled with bone specific markers: antiosteocalcin (OCN), anti-osteopontin (OPN) and anti-bone morphogenetic protein 2 (BMP2) antibodies. Hoechst was used to visualize the nucleus. Confocal images were acquired using Zeiss LSM800 Airyscan 40x oil immersion objective lens as described earlier. Western immunoblot analysis was performed for the quantitative estimation of the OPN protein generated over different scaffold types at day 11 p.s. The hFob cells were lysed in ice-cold lysis buffer containing Triton X-100 (1%) and protease inhibitors. Detergent-insoluble material and nuclei were removed by centrifugation for 5 min at 10000 rpm. The protein content of the cell lysates was determined by the Bio-Rad Protein Assay. Equal amount of cell lysates (20 g each) were resolved on SDS-PAGE gels and subsequently transferred onto polyvinylidene difluoride (PVDF) membranes. After blocking in 5% milk powder in TBS-0.05% Tween 20 (TBST) for 1 h at room temperature, the membranes were washed three times in TBST. The membranes were then incubated overnight at 4°C with diluted primary antibody in 5% milk with gentle rocking. After three washes in TBST, the membranes were incubated with HRP-conjugated secondary antibody for 1 h at room temperature. After three further washes in TBST, the membranes were incubated with the LumiGLO® chemiluminescent detection system (Cell Signalling Technology) and exposed to light sensitive film for various times. Densitometric analyses of the Western immunoblots were performed using ImageJ software. 2.13. Statistical Analysis. All the cell culture experiments were run in triplicates at various time points specified and the data expressed as mean  standard deviation (SD). Statistical differences were determined using a one-way ANOVA with a Turkey adjustment for multiple comparisons. Difference was considered statistically significant at p  0.05. 3. RESULTS AND DISCUSSION 3.1.

Morphological Analysis

of Electrospun Collagen

Nanofibrous Scaffolds.

Morphological features of ES collagen scaffolds with or without catecholamines and catecholamines/Ca2+ were investigated by electron microscopy. Scanning Electron 10

ACCEPTED MANUSCRIPT Microscopy (SEM) of pristine collagen mats electrospun from a 10% dope solution in HFIP revealed the presence of smooth, bead-free and uniform fibre morphology with an average diameter () of 900 ±150 nm (Figure 2a). Addition of 10% (w/w) catecholamines caused two noteworthy morphological changes in the electrospun collagen mats: i) a significant decrease (~3 folds) in the -values (331 ± 46 nm for COLL_DA and 323 ± 43 nm COLL_NE) and ii) presence of catecholamines merge the two fibers and remove the identity of individual fibers at the contact points, thus forming welded or soldered junctions (Figure 2b, c and insets at top right). Since the electrospinning was performed in a poorly hydrogen bonded solvent which promotes intramolecular hydrogen bonding, the presence of higher concentration of catecholamines at the contact points might promote the inter-fibre adhesion, thus forming welded junctions.29,30 Previous studies suggested that electrospun collagen mats prepared from fluoro alcohols lack adequate mechanical strength and aqueous stability, therefore may not be suitable for oxidative polymerization of catecholamines by conventional alkaline Tris-HCl (pH 8.5) method.31,32 To avoid the deleterious effect of alkaline solution on collagen nanofibers, we exposed the catecholamine-loaded mats to (NH4)2CO3 in order to induce the oxidative polymerization of catecholamines. Henceforth, this method is referred as ammonium carbonate diffusion method (ADM). Since the entire reaction was carried out at solid-vapor interface, one would expect the morphological features of electrospun collagen remain intact. Indeed, substantial increase in the formation of welded junctions was observed after exposure of the catecholamines-loaded mats to (NH4)2CO3 (Figure 2d, e). The effect was remarkable in the case of NE containing mats, as numerous inter-fibre adhesion and welded junctions resulted in discontinuous porous coatings after ADM (Figure 2e). Brown coloration of the scaffolds after ADM further confirmed the formation of oxidative products of catecholamines (Insets at bottom left of Figure 2d, e). Notably, addition of 20 mM Ca2+ to collagen-catecholamine dope solution dramatically increased the formation of welded junctions in comparison to mats electrospun without Ca2+ (Figure 2f, g and insets at top right). The color of the mats changed from white to brown (DA)/pale pink (NE) upon addition of 20 mM Ca2+ (Figure 2f, g and insets at bottom left). These results suggest possible cation-induced electrochemical oxidation of catecholamines to polycatecholamines that possess inherent adhesive properties, thus

11

ACCEPTED MANUSCRIPT promoting inter fibre adhesion.33 As the color of the dope solution containing catecholamines and Ca2+ did not change prior to electrospinning, it is likely that the brown coloration of the mats could be attributed to the electrochemical potential-induced oxidation of DA and NE during the electrospinning process.34,35 After ADM, the catecholamines-Ca2+ loaded collagen mats displayed more intense coloration without profound changes in the morphology (Figure 2h, i). High-resolution TEM images showed distribution of sub-nm sized nanoparticles along the length of individual fibers, indicating the formation of CaCO3 (inset at top right of Figure 2h, i). The appearance of diffused rings in the selected area diffraction pattern confirmed the amorphous nature of the particles (Figure S1). Detailed mineralogical characterization of the particles was difficult as we could not detect clear signals by FTIR, Raman or XRD methods. Nevertheless, taking advantage of the formation of gaseous ammonia and CO2 upon decomposition of (NH4)2CO3, we reported a simple strategy to carry out simultaneous crosslinking and mineralization of ES collagen fibers.

12

ACCEPTED MANUSCRIPT

a) a)

b) b)

c) c)

d) d)

e) e)

After

(NH4)2CO3 Exposure

Before

+DA or NE and No Ca2+

f)f)

g) g)

h) h)

i)i)

After

(NH4)2CO3 Exposure

Before

+DA or NE and 20mM Ca2+

Figure 2. Morphological analysis of electrospun mats prepared under various conditions. (a) pristine collagen mats. (b) and (c) collagen mats containing 10% DA and NE, respectively. (d) and (e) mats shown in (b) and (c) after exposure to (NH4)2CO3. Scale bar = 1 µm. Photographs of the mats are shown in the bottom panel. Note the increased amount of welded junctions and coloration after ADM in both the catecholamines loaded mats. Effect of CaCl2 on the morphology of electrospun collagen containing (f) DA and (g) NE. Note the complete 13

ACCEPTED MANUSCRIPT coating of polycatecholamines along the entire surface of the mats as well as significant color changes in the mats. (h) and (i) Morphology of electrospun mats in (f) and (g) after exposure to (NH4)2CO3. Note the intense brown coloration of the mats containing DA. Insets in (h) and (i) are high resolution TEM images displaying the formation of CaCO3 particles after ADM. 3.2. Characterization of Collagen Composites by XPS. We next focused our attention on the characterization of composite structures formed. To infer the changes in chemical bonding environments around carbon and nitrogen of collagen fibers upon incorporation of catecholamines (DA and NE) with CaCl2 followed by ADM, we performed the XPS measurements. The shape of high resolution C 1s and N 1s core-level spectra indicated the presence of multiple bonding components in the ES mats (Figure 3). Deconvolution of C 1s spectra revealed the presence of four peaks in each sample; namely C1, C2, C3 and C4 which correspond to C-C/C-H, C-N, C-O and C=O bonding, respectively (Figure 3a-j). Similarly, the deconvolution of N 1s spectra revealed the presence of two peaks in each sample; namely N1 and N2 which are assigned to R2NH and RNH2 bonding, respectively. The peak positions for these bonding are provided in Table S1 and are found to be in good agreement with the reported literature.36,37 Furthermore, an area ratio method was used to estimate the bonding content of each peak for all samples and the results are summarized in Table 1. C 1s spectra of Coll_DA_Ca and Coll_NE_Ca samples revealed significant increase in C-N bonding intensity and atomic composition, corroborating the color changes observed during electrospinning. The increase in C-N bonding intensity was further more pronounced after ADM.

14

ACCEPTED MANUSCRIPT C 1s C2

C3 C4

(b)

ES_Coll

N2

(b)

Intensity (a.u.)

Intensity (a.u.)

(b)

(f)

Coll_DA_Ca

(g)

Coll_NE_Ca

Coll_pDA_Ca

Coll_pNE_Ca

345

(l) 2p1/2

ES_Coll

Intensity (a.u.)

(h) Coll_DA_Ca

Coll_NE_Ca

Coll_pDA_Ca

(i)

Coll_pNE_Ca

195

200

205

210

Binding Energy (eV)

(m)

O 1s ES_Coll

Coll_DA_Ca

Intensity (a.u.)

(e)

Intensity (a.u.)

Intensity (a.u.)

(e)

(e)

Cl 2p

(c)

Intensity (a.u.)

Intensity (a.u.)

(c)

(d)

355

(d)

Intensity (a.u.)

Intensity (a.u.)

(c)

350

Binding Energy (eV)

2p3/2 (d)

Ca 2p 2p1/2

2p3/2

N1

(a)

Intensity (a.u.)

(a)

C1

Intensity (a.u.)

Intensity (a.u.)

(a)

(k)

N 1s

(j)

Coll_NE_Ca

Coll_pDA_Ca

Coll_pNE_Ca

280 282 284 286 288 290 292 396

Binding Energy (eV)

398

400

402

404

Binding Energy (eV)

528

532

536

540

Binding Energy (eV)

Figure 3. XPS characterization of electrospun mats prepared under various conditions. High resolution C 1s spectra of (a) ES_Coll, (b) Coll_DA_Ca, (c) Coll_NE_Ca, (d) Coll_pDA_Ca and (e) Coll_pNE_Ca. High resolution N 1s spectra of (f) ES_Coll, (g) Coll_DA_Ca, (h) Coll_NE_Ca, (i) Coll_pDA_Ca and (j) Coll_pNE_Ca. (k) – (m) show the high resolution Ca 2p, Cl 2p and O 1s spectra for the electrospun mats. To probe the modifications of bonding environments around calcium ions, Ca 2p and Cl 2p core level spectra were also recorded for all samples (Figures 3k, l). For Coll_DA_Ca 15

ACCEPTED MANUSCRIPT and Coll_NE_Ca samples, the Ca 2p3/2 peak at 348.0 eV, Cl 2p3/2 peak at 198.3 eV and Cl 2p1/2 peak at 199.85 eV indicate that the bonding environments of calcium or chloride ions did not change after electrospinning.38-40 However, two important observations confirmed the formation of CaCO3 in the mats after ADM. First, the intensity of Ca and Cl peaks decreased in both Coll_pDA_Ca and Coll_pNE_Ca mats, though the effect was substantial in the mats containing dopamine. Second, the Ca 2p3/2 (347.6 ± 0.05 eV) and the Cl 2p3/2 (at 198.0 eV) peaks shifted towards lower binding energy, when compared to Coll_DA_Ca or Coll_NE_Ca mats. The appearance of Ca 2p3/2 in the range 346.5 - 347.9 eV have been reported for various polymorphs of CaCO3.41-43 Table 1. Quantitative analysis of various bonds examined from C1s and N1s core level spectra. Samples

ES_Coll Coll_DA_Ca Coll_NE_Ca Coll_pDA_Ca Coll_pNE_Ca

Percentage of the Constituent Peaks C1s Core Level C-C/C-H (%) C-N (%) C-O (%) 45.6 24.5 3.9 46.7 25 6 48.3 25 4.7 38.8 33.5 1.1 42.6 32 1.3

C=O (%) 26 22.3 22 26.6 24.1

To obtain further insight into the formation of CaCO3 in the mats after ADM, O 1s core level spectra were examined for the cross-linked nanofibers. A weak up shift in the position of O 1s peak from 530.8 eV to 531.0 eV was observed ES_Coll upon incorporation of DA/NE and Ca2+ i.e., for Coll_DA/NE_Ca mats (Figure 3m). Since Ca 2p and Cl 2p spectra revealed no apparent change in the spectra for Coll_DA/NE_Ca mats, the marginal up shifting of O 1s peak in these samples could be due to weak interactions of Ca 2+ with catecholamines. However, after ADM, the O 1s peak was further up shifted to 531.2 eV for Coll_pDA/pNE_Ca mats, confirming the formation of carbonate structures in these mats. 39-41 Together with TEM results, these observations confirmed the formation of CaCO3 in the mats after (NH4)2CO3 treatment. 3.3. Wettability of Collagen Composites. The surface wetting properties of the ES mats were assessed by measuring the time-dependent changes in the advancing contact angle with water. For pristine collagen mats (ES_Coll), the water contact angle (WCA) reached up to 59.3±1.4 in 60 seconds from an initial value of 74.4±0.1 (Figure 4a). Incorporation of DA (Coll_DA), decreased the initial WCA by about 20, indicating increased wettability of the mats and reached a final value of 40.6±0.1 after 1 min. After ADM treatment (Coll_pDA), 16

ACCEPTED MANUSCRIPT the initial WCA decreased sharply from 92.1±4.5 to a final value of 52.5±0.1.

The

presence of Ca2+ in the dope solution (Coll_DA_Ca) has dramatic effect on the initial WCA and the values decreased exponentially and reached a value of 76.7±3.9. Interestingly, after ADM exposure, no apparent change in the WCA was observed for the Coll_pDA_Ca. To shed further insight into the WCA measurements, we determined the static WCA (static), by extrapolation of the linear part of the curve to zero time. The results indicated that collagen mats containing DA before or after ADM decreased static when compared to pristine collagen mats (Figure 4b). In contrast, the values increased significantly for Coll_DA_Ca mats and remained unaltered after ADM (Figure 4b). These results corroborate our earlier observations that the presence of Ca2+ together with DA could trigger the oxidative polymerization during electrospinning, thus producing mats with surface wettability that was similar to Coll_pDA mats. In the case of ES collagen mats containing NE, a similar trend in the

a)

150

****

90

C o n ta c t A n g le , 

b)

***

time-

C o ll_ D A _ C a

 s ta tic (  )

E S _ C o ll C o ll_ D A

C o ll_ p D A C o ll_ p D A _ C a

C _

N

E

p _

N

ll

_

C

p

o

_ ll

ll

o

C

o

C

T im e , s e c o n d s

a

E

a C _ N

o

E

ll

80

C

60

_

40

S

20

E

0

_

C

N

o

E

ll

0

0

C o n ta c t A n g le , 

E S _ C o ll C o ll_ N E C o ll_ N E _ C a

d)

* ****

90

 s ta tic (  )

c)

120

C o ll_ p N E C o ll_ p N E _ C a a C _

D

A

p

D

_ ll

_

p

o C

o

ll

C

_ ll o C

T im e , s e c o n d s

A

a C _ A D

ll o

80

C

60

_

40

S

20

E

0

_

C

D

o

A

ll

0

0

dependent WCA (Figure 4c) and static were observed (Figure 4d).

Figure 4. Surface wettability of electrospun collagen mats prepared under various conditions. (a) and (c) show the time-dependent changes in the advancing contact angle for mats containing DA and NE, respectively. The symbols indicate average data points at every 10 seconds and solid or broken lines represent the model fit. (b) and (d) compares the static 17

ACCEPTED MANUSCRIPT water contact angle (static) determined from the dynamic contact angle measurements. * p≤0.05, **p<0.01, ***p<0.001 and ****p<0.0001 compared to pristine collagen scaffold by ttest or 1-way ANOVA. 3.4. Mechanical Properties of Collagen Composites. Next we investigated the effect of incorporating catecholamines/Ca2+ and subsequent mineralization and crosslinking on the mechanical properties of collagen mats. Typical stress-strain curves for various collagen mats measured by uniaxial tensile testing are shown in Figure 5a and b. We determined peak stress (), elongation at break (b), Young’s modulus (E’) and work of failure (Jlc) from the stress-strain curves (Table 2). Detailed statistical analyses of the mechanical properties for various mats are presented in Table S2. For ES_Coll, a brittle-like behavior was observed with a  value of 4.9±0.5 MPa and b of 6.0±1.3%. A tensile modulus of 156.7±29 MPa and mechanical toughness of 0.22 ± 0.07 MJm-3 were estimated from the stress-strain curves (Figure 5a). ES_Coll Coll_DA Coll_pDA Coll_DA_Ca Coll_pDA_Ca

Stress [MPa]

25 20

a)

b)

70 ES_Coll Coll_NE Coll_pNE Coll_NE_Ca Coll-pNE_Ca

60

Stress [MPa]

30

15 10

50 40 30 20

5

10 0

0 0

10

20

30

40

0

50

2

4

6

1000

C

60

Co

DA Co

ll_

_ DA

Ca

Co

ll_

p

_ DA

Ca

0

1

CO

LL

C

l_ ol

DA Co

ll_

pD

A

Co

ll_

_ DA

Ca

Co

ll_

p

_ DA

Ca

-N

E CO

LL

pN

E

CO

LL

E -N

-C

CO

a LL

-

E pN

-C

8

b J lc

0

0

a

e)

d)

3

0 LL

l

M J /m

10

MPa

100

O _C

E

ol

ES

_

W o r k o f F a ilu r e ( J l c )

1000

Y o u n g 's M o d u lu s ( E ') ,

P e a k S tre s s ( ), M P a

c)

0 C S_

50

 E'

ES

J lc

( b), %

_

p ll_

5

b

1 DA

E lo n g a t io n a t B r e a k

ES

50

3

0 l_ ol

14

( b), %

MPa 10

LL

12

M J /m

P e a k S tre s s ( ), M P a

100

CO

10

W o r k o f F a ilu r e ( J l c )

E'

Y o u n g 's M o d u lu s ( E ') ,



E lo n g a t io n a t B r e a k

50

8

Srain [%]

Strain [%]

CO

LL CO

LL

NE CO

L

p L-

NE CO

LL

-

NE

Ca

CO

LL

-p

NE

Ca

f)

Figure 5. Mechanical properties of electrospun collagen mats. Stress – strain curves of the electrospun mats prepared under various conditions containing (a) DA and (b) NE. Peak stress () and tensile stiffness (E’) values determined from the curves are shown in bar graphs for the mats containing (c) DA and (d) NE. E’ is shown in log10 units for a better 18

ACCEPTED MANUSCRIPT comparison. The increase in elongation at break (b) and toughness (Jlc) are shown for mats containing (e) DA and (f) NE. Note the marked increase in elasticity and stiffness of mineralized collagen mats crosslinked with polycatecholamines. Addition of 10% DA, increased the elastic properties of the mats as indicated by an increase in b and Jlc values (Figure 5c). After ADM, a brittle-like behavior was observed as indicated by significant increase in  (15.1±3.8 MPa) or E’ (512.6±136.9 MPa) and low b values (4.1±0.6%, Figure 5c, d). However, incorporation of Ca2+ into the dope solution containing DA had profound influence on tensile properties, as the mats prepared from the solution displayed remarkable increase in , E’ and b values (Figure 5c, d). It was interesting to note that the tensile properties of Coll_DA_Ca approached the values obtained for Coll_pDA, confirming that the presence of Ca2+ in the dope solution triggered the oxidative polymerization of DA. For Coll_pDA and Coll_pDA_Ca no significant difference in  and E’ values (p>0.05) were observed whereas considerable differences were apparent in the elastic properties. These results suggest that the formation CaCO3 increased the toughness of the composites without affecting tensile stiffness and strength of polycatecholamines crosslinked collagen. A similar trend was observed in the case of mats containing NE (Figure 5b, e and f). In fact, the influence of Ca2+ on the mechanical properties of the mats containing NE was so remarkable that E’ values approached sub-GPa with higher elasticity/toughness than observed for mats containing DA (Table 2). These results suggest that incorporation of Ca2+ into catecholamines containing collagen and subsequent mineralization increased the elastic properties with concomitant enhancement in mechanical strength and stiffness. Such remarkable enhancement in both stiffness and toughness of the composites compared to pristine polymers could not be achievable from the composite mats prepared by direct mixing of minerals and polymers reported by others.12,17-20,44,45

19

ACCEPTED MANUSCRIPT Table 2. Mechanical properties of electrospun collagen mats prepared under various conditions. Significance values: *, p≤0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001 and ns, p>0.05 by t- test or 1-way ANOVA. Sample

Peak Stress (MPa)

Failure Strain (%)

Young’s Modulus (MPa)

Work of Failure (MJ/m3)

ES_Coll

4.9±0.5

6.0±1.3

156.7±29

0.22±0.07

Coll_DA

6.3±0.5ns

19.2±4.4**

108.8±27.3ns

0.93±0.44**

Coll_pDA

15.1±3.8*

4.1±0.6ns

512.6±136.9**

0.334±0.03ns

Coll_DA_Ca

12.8

28.1

359.6

3.9

Coll_pDA_Ca

21.7±5.7***

15.1±4.1*

675.8±81.2****

0.83±0.00*

Coll_NE

3.36±0.5ns

4.5±2.3ns

88.4±36.1ns

0.07±0.03ns

Coll_pNE

11.8±3.6 ns

4.5±1.6ns

400.5±126ns

0.32±0.25ns

Coll_NE_Ca

29.9±6.9****

12.4±1.8**

1065±240****

3.0±0.8****

Coll_pNE_Ca

37.6±20.3****

22.2±20****

1070±367***

5.36±2.3****

To shed more insight into the mechanical properties, we imaged the fracture surface of various mats by SEM. The deformed structure of ES_Coll and Coll_DA and Coll_NE mats contained numerous collagen fibrils (Figure 6a-c). The deformed surface of Coll_pDA contained numerous fracture steps indicating weak resistance to the applied stress (Figure 6d), confirming the brittle-like behavior.46 For Coll_pNE due to the formation of welded junctions and smooth polycatecholamine coating, the fractured surface appeared as fibrereinforced composite structures (Figure 6e). However, incorporation of Ca2+ considerably altered the fracture morphology as detected by significant increase in crack branching and river-like lines on the fractured surface of Coll_DA_Ca and Coll_NE_Ca mats (Figure 6f, g). The roughness of the surface increased significantly for Coll_pDA_Ca and Coll_pNE_Ca mats (Figure 6h, i). The high abundance of welded junctions and coating formed in the presence of Ca2+ may account for the increase in the mechanical properties of the mats while the formation CaCO3 particles increase the roughness and prevents the crack propagation thus enhancing the toughness of the composites.46

20

ACCEPTED MANUSCRIPT

a)

b)

c)

d)

e)

After

(NH4)2CO3 Exposure

Before

+DA or NE and No Ca 2+

f)

g)

h)

i)

After

(NH4)2CO3 Exposure

Before

+DA or NE and No Ca2+

21

ACCEPTED MANUSCRIPT Figure 6. Fracture morphology of collagen mats. SEM images of electrospun mats after tensile testing. (a) ES_Coll, (b) Coll_DA, (c) Coll_NE, (d) Coll_pDA, (e) Coll_pNE, (f) Coll_DA_Ca, (g) Coll_NE_Ca, (h) Coll_pDA_Ca and (i) Coll_pNE_Ca. Scale bar = 1 µm. Note that the mineralized collagen mats (h) and (i) displayed extensive roughness and corrugation indicating considerable increase in tensile strength, stiffness and toughness. 3.5. Photoluminescence Properties of ES Collagen Mats. Recent studies have shown that the addition of cationic polyethylenimine (PEI) during alkaline oxidative polymerization resulted in PEI-pDA copolymer with enhanced fluorescence emission properties.47 Therefore, we investigated the optical properties of catecholamines crosslinked mats by fluorescence microscopy and spectroscopy. Pristine ES_Coll mats displayed a weak blue/green emission (405/488 nm) whereas mats containing polycatecholamines displayed intense blue (405 nm) and green (488 nm) emission (Figure 7a - c). In particular, the fluorescence intensity was stronger at the soldered junctions. For collagen mats containing dopamine, substantial decrease in both blue (405 nm) and green (488 nm) fluorescence intensities was observed upon mineralization in Coll_pDA_Ca fibres (Figure 7b, d). However, for mats containing NE, mineralization did not alter the blue (405 nm) fluorescence intensity whereas a slight increase in the green (488 nm) fluorescence intensity was observed for Coll_pNE_Ca fibres (Figure 7c, e). Quantitative estimation of the average fluorescence intensities further corroborated the above results (Figure 7f). The origin of differences in the photoluminescent properties remain unclear, it is likely that the two polycatecholamines may interact differently with the inorganic minerals formed after ADM. The development of composite structures with excellent mechanical, photoluminescent and biological (see below) properties can function as non-invasive real-time probes as well as for quantifying the scaffold degradation patterns and tissue integration without sacrificing the animals 48-50 and would open a new paradigm for the implementation of smart multifunctional scaffolds for tissue engineering and regenerative medicine.

22

ACCEPTED MANUSCRIPT 405 nm

488 nm

Merge

ES_Coll

a)

Coll_DA_Ca

b)

Coll_NE_Ca

c)

Coll_pDA_Ca

d)

Coll_pNE_Ca

e)

f)

ns

25000

**

****

RFU

****

0 _ ES

Co

ll

Co

ll_

_ DA

Ca Co

ll_

_ NE

C

a

Co

ll_

A pD

_C

a

Co

ll_

pN

E_

C

a

Figure 7. Confocal fluorescence images showing the photoluminescent properties of collagen mats. (a) ES_Coll, (b) Coll_DA_Ca, (c) Coll_NE_Ca, (d) Coll_pDA_Ca and (e) Coll_pNE_Ca. The excitation wavelengths used are 405 nm (blue) and 488 nm (green). Scale bar = 20 m. At least 20 microscopic fields were visualized from 3 different mat preparations and representative images are shown. Images were acquired by Zeiss LSM800 Airyscan using ×63 oil immersion objective lens. (f) Average blue and green fluorescence intensities (RFU) obtained from various ES collagen mats (mean ± S.D.). 3.6. Biological Properties of Collagen Composites. Next, we examined the biological properties of the ES collagen scaffolds by assessing their cytocompatibility using hFob cells in terms of their ability to enhance cell adhesion, proliferation, mineralization and expression of key osteogenic protein markers. Cells were seeded on ES_Coll, Coll_DA_Ca, Coll 23

ACCEPTED MANUSCRIPT _NE_Ca, Coll _pDA_Ca and Coll _pNE_Ca mats. Cells seeded on the tissue culture plate (TCP) served as control. After 3, 6 and 9 days post seeding (p.s.) of hFob on various scaffolds, cells were stained with CMFDA and Cytiva Cell Health Reagents to obtain semiquantitative information about live/dead cells. All the scaffolds displayed no discernible cytotoxicity to hFob cells as the % live cells remained >90% at various time points (Figure 8f), thus confirming the excellent biocompatibility of scaffolds for hFob. Confocal images of the cells cultured on various scaffolds are shown in Figure 8a-e at three different time points. Interestingly, an increased cell populations and spreading was observed in the Coll_pDA/pNE_Ca mats (Figure 8d, e), as indicated by increased staining of CMFDA and nuclei (blue) after 9 days p.s. These results suggest relatively more cells spreading on Coll_pDA/pNE_Ca mats than on ES_Coll. Confocal z-stack imaging allowed us to examine the level of cell penetration into various collagen mats. A time-dependent increase in the cell penetration was observed on all the mats with markedly increased penetration in Coll_pDA/pNE_Ca mats. Together with xy-scans, these results suggest increased cell spreading and active cell migration on mineralized scaffolds. To obtain better insight about the cell penetration into scaffolds, hFob cells were stained with Alexa Fluor 647-Phalloidin and Hoechst dyes and imaged by confocal microscopy (Figure 9). Images clearly showed an enhanced cell growth on Coll_pDA_Ca and Coll_pNE_Ca mats, consistent with the previous results. Confocal z-sections of the images further confirmed a time-dependent increase in cell penetration with multiple cell layers on all ES scaffolds (Figure 9). Moreover, enhanced hFob cell penetration was observed in mineralized mats (Coll_pDA_Ca and Coll_pNE_Ca) with a maximum depth of ~28m at 9 days p.s. (Figure 9d and e, top panels). To confirm if the collagen mats remained intact after immersion in cell culture, we imaged the scaffolds by taking advantage of the fluorescence properties of collagen. For all the ES mats, the fibre morphology remained intact indicating no significant degradation occurred in the presence of cell culture physiological environment, despite the increase in cell proliferation and differentiation (Figure S2).

24

ACCEPTED MANUSCRIPT 3 days

6 days

9 days

a) ES_Coll

Z, 28 μm

b) Coll_DA_Ca

Z, 28 μm

c) Coll_NE_Ca

Z, 28 μm

d) Coll_pDA_Ca

Z, 28 μm

e) Coll_pNE_Ca

Z, 28 μm

f)1 5 0 3 d a y s p .s .

6 d a y s p .s .

L e th a lity ( % )

C e ll V ia b ility /

9 d a y s p .s .

0

TC

P _ ES

Co Co

ll ll

A _D

_C Co

a ll

E _N

_C

Co

a

ll_

A pD

_C

Co

a

ll_

E pN

_C

a

Figure 8. Cytocompatibility of collagen mats. Confocal fluorescence images of hFob cells cultured on various collagen mats stained with CMFDA after 3, 6 and 9 days p.s. (a) ES_Coll, (b) Coll_DA_Ca, (c) Coll_NE_Ca, (d) Coll_pDA_Ca or (e) Coll_pNE_Ca. Images were acquired by Zeiss LSM800 Airyscan microscope using ×63 oil immersion objective lens. The top panels in each confocal images show the merged 28 µm z-stack sections. Scale 25

ACCEPTED MANUSCRIPT bar = 20 µm. (f) hFob cell viability (quantified from live/dead cell ratio) cultured on various scaffolds and TCP. Mean  SD, n = 3 3 days

6 days

a)

9 days

ES_Coll

Z, 28 μm

b)

Coll_DA_Ca

Z, 28 μm

c)

Coll_NE_Ca

Z, 28 μm

d)

Coll_pDA_Ca

Z, 28 μm

e)

Coll_pNE_Ca

Z, 28 μm

Figure 9. Morphological analysis of hFob cells seeded on various collagen mats. Cells growing on a) ES_Coll, (b) Coll_DA_Ca, (c) Coll_NE_Ca, (c) Coll_pDA_Ca, or (e) Coll_pNE_Ca mats were fixed on 3, 6, or 9 days p.s., stained with Alexa Fluor 647Phalloidin (to visualize cellular morphologies, violet) and Hoechst (to visualize nuclei, blue). Confocal fluorescence images were acquired by Zeiss LSM800 Airyscan microscope using 26

ACCEPTED MANUSCRIPT ×63 oil immersion objective lens. Collagen mats exhibiting green fluorescence can also be seen in the images. Scale bar = 20 µm. Confocal z-stack images (28 m sections) of individual samples (top panel on each confocal images) are given. Metabolic activity of seeded cells on various scaffolds was monitored by MTS assay after 3, 6, 9, 11 and 14 days p.s to determine the proliferation of hFob cells. To obtain a better insight, data was expressed in cell proliferation ratio after dividing cell number at each time point with initial cell density. As shown in Figure 10a, all the ES scaffolds supported cell attachment and proliferation until 9 days p.s., indicated by the increase in cell proliferation values, and levelled off after 11 days p.s. Since collagen provides active cell recognition sites for hFob adhesion and proliferation, we observed higher rate of cell proliferation on ES scaffolds than TCP. Among the various collagen mats, no statistically significant differences in hFob proliferation were observed at 3 days p.s. (Table S3). However, catecholamines loaded mineralized mats (Coll_pDA/pNE_Ca), displayed increased proliferation as early as 6 days p.s. Cells cultured on all the catecholamine/Ca2+ mats (Coll_DA/NE_Ca and Coll_pDA/pNE_Ca) showed significantly pronounced time dependant metabolic activity compared to ES_Coll (Figure S3a). Among the two catecholamines, NE-loaded mats promoted higher cell proliferation than mats containing DA. Analysis of CMFDA and MTS results revealed the following conclusions: i) Compared to TCP, a unanimous increase in the hFob proliferation was observed for the mats at all the time points. ii) Incorporation of Ca2+ into the catecholamines loaded collagen and subsequent exposure to (NH4)2CO3 promoted enhanced cell spreading and proliferation, iii) Among the two catecholamines, cell proliferation was more pronounced in the NE incorporated mats i.e., COLL_NE_Ca (p < 0.0001 vs. Coll_DA_Ca at day 11 p.s.) and COLL_pNE_Ca (p < 0.0001 Coll_pDA_Ca day 11 p.s.). These results suggest that the polycatechoalmine-CaCO3 composite structures are non-cytotoxic for the hFob and stimulate the adherence and proliferation of bone cells and these properties can be modulated by appropriate choice of catecholamines.

27

ACCEPTED MANUSCRIPT

TCP E S _ C o ll

C o ll _ p D A _ C a

a)

0 .1 5

C o l l_ p N E _ C a

A L P A c tiv ity /C e ll N u m b e r

20

C e ll P r o life r a tio n

C o ll _ D A _ C a

15

C o ll_ N E _ C a

10

5

TCP

C o ll _ p D A _ C a

E S _ C o ll

C o l l_ p N E _ C a

b)

C o l l_ D A _ C a C o ll_ N E _ C a

0 .1 0

0 .0 5

0 .0 0

0 3 Days

6 Days

9 Days

11 D ays 14 D ays

3 Days

6 Days

9 Days

11 D ays 14 D ays

Figure 10. hFob cell proliferation and ALP activity on various composite collagen mats. (a) Metabolic activity of hFob cells assessed by MTS assay at various time points. Data is reported in term of cell proliferation estimated by dividing cell number at each time point with initial cell density and presented as mean  standard deviation (n = 5). (b) Intracellular ALP activity of hFob cells assessed by pNPP assay at various time points. ALP activity was normalized by the cell number (mol p-Nitrophenol of hFob cells/h/cell number) and reported. Data are reported as mean  SD (n = 3). Note the marked increase in osteoblast proliferation and differentiation on mineralized mats containing polycatecholamines. ALP is a key component of bone matrix vesicles that catalyzes the cleavage of organic phosphate esters, plays a significant role in the formation of bone mineral and is an early indicator of immature osteoblast activity.51 ALP activity is also a marker of early osteoblastic differentiation and commitment of the stem cells towards the osteoblastic phenotype.52 We, therefore, examined the ability of the mats to promote osteogenic differentiation by measuring cellular ALP activity, on 3, 6, 9, 11 and 14 days p.s. The ALP activity was normalized by cell number and shown in Figure 10b. No significant variation in the ALP activity was observed for all the investigated collagen scaffolds on 3 days p.s. (Figure 10b and Table S4). However, compared to the pristine collagen or TCP, the ALP activity increased dramatically for all the catecholamine/Ca2+ incorporated scaffolds (p< 0.0001) after 6 day p.s. of hFob. The hFob differentiated stably for an extended period of time when cultured on Coll_pDA/pNE_Ca mats after 6-14 days p.s. than cells cultured on Coll_DA/NE_Ca, suggesting that mineralized scaffolds promote cell differentiation as well (Figure S3b). Among the two catecholamines, a significantly higher ALP activity was observed for cells cultured on Coll_pNE_Ca than on Coll_pDA_Ca mats during early stages (Table S4), confirming increased osteogenic differentiation potential of oxidative products of norepinephrine. Together with the cell proliferation assays, these results demonstrate that Coll_pDA/pNE_Ca mats are superior in terms of osteoblasts cell growth, penetration and differentiation.

28

ACCEPTED MANUSCRIPT Upon osteoblast differentiation, the hFob cells enter into the mineralization phase to deposit the mineralized ECM. The capacity of hFob to deposit minerals is a marker for osteogenic efficiency and can be monitored by ARS staining of the cells cultured on different scaffolds after 9 days p.s. Figure 11 shows the optical images of various samples stained with ARS, wherein the bright red staining indicates the calcium mineralization due to ARS binding. When compared to TCP, all the ES mats displayed substantial ARS staining, indicating increased mineralization of the scaffolds. ARS staining was enhanced in the mats containing catecholamine/Ca2+and more pronounced in the polycatecholamines-CaCO3 composite mats than in ES_Coll, consistent with the increased ALP activity observed on these scaffolds. ARS stained optical images further showed thick bone nodule formation on hFob cells cultured on Coll_DA/NE_Ca or Coll_pDA/pNE_Ca mats (Fig. 11c-f). These results suggest that the biochemical cues arising from polycatecholamines-CaCO3 composite structures stimulate the hFob adhesion, proliferation and differentiation. Quantitative estimation of the colorant on various mats by cetylpyridinium chloride assay at 14 days p.s. further confirming the increased osteogenic potential of Coll_pDA/pNE_Ca (Figure 11g). Similar to the trend observed with the ALP activity, higher calcium nodules (p<0.0001) were observed for Coll_pNE_Ca than Coll_pDA_Ca nanofibrous scaffolds (Table S5).

29

ACCEPTED MANUSCRIPT

a)

b)

c)

d)

e)

f)

g)

A b s o rb a n c e 5 4 0

nm

2

0 _ ES

Co C

ll

l_ ol

DA

_C C

a

l_ ol

NE

_C

Co

a

p ll_

DA

_C

Co

a

p ll_

NE

_C

a

TC

P

Figure 11. ARS staining of calcium deposition on collagen mats. Determination of calcium deposition on various collagen mats by ARS staining. (a) TCP, (b) ES_Coll, (c) Coll_DA_Ca, (d) Coll_NE_Ca, (e) Coll_pDA_Ca and (f) Coll_pNE_Ca. Scale bar = 50 m. (g) Quantitative estimation of calcium deposition by cetyl pridinium bromide assay at 14 days p.s. Data are reported as mean ± SD from three independent experiments. 30

ACCEPTED MANUSCRIPT b) ES_Coll

c) Coll_DA_Ca d) Coll_NE_Ca e) Coll_pDA_Ca f)

Coll_pNE_Ca

a) TCP

b) ES_Coll

c) Coll_DA_Ca d) Coll_NE_Ca e) Coll_pDA_Ca f)

Coll_pNE_Ca

a) TCP

b) ES_Coll

c) Coll_DA_Ca d) Coll_NE_Ca e) Coll_pDA_Ca f)

Coll_pNE_Ca

BMP-2

OCN

OPN

a) TCP

g)

OPN

2 .5

( O P N /G A P D H )

R e la t iv e P r o t e in E x p r e s s io n

GAPDH

0 .0 TC

P ES

_C C

ol

l

l_ ol

DA

_C

Co

a

p ll_

DA

_C C

a

l_ ol

NE

_C

Co

a

p ll_

NE

_C

a

Figure 12. Expression of key osteogenic proteins by hFob cells seeded on various collagen mats. Representative immuno staining images showing the expression of osteocalcin (OCN), osteopontin (OPN) and bone matrix protein 2 (BMP-2) after 11 days p.s. (a) TCP, (b) ES_Coll, (c) Coll_DA_Ca, (d) Coll_NE_Ca, (e) Coll_pDA_Ca and (f) Coll_pNE_Ca. Scale bar = 20 m. (g) Western immunoblot analysis of OPN expression in hFob cells cultured on ES mats. Blots were re-probed with anti-GAPDH as loading control, and densitometry quantification of relative OPN expression is shown. Results represent mean±SD of three independent experiments. To obtain further insight into the biochemical cues on scaffolds-cell interactions, we determined the expression of key osteo-specific marker proteins such as osteocalcin (OCN), osteopontin (OPN) and bone morphogenetic protein 2 (BMP-2) by immunofluorescent labelling (Figure 12). As observed in MTS/ALP assays and in ARS staining, substantial amount of immunofluorescence signals for all the osteo-specific marker proteins (OCN, OPN and BMP-2) was observed when hFob cells were cultured on the ES fibres. (Figure 12). The osteogenic marker proteins OPN, OCN and BMP-2 were increased in Coll_pDA_Ca and 31

ACCEPTED MANUSCRIPT Coll_pNE_Ca than in ES_Coll or Coll_DA/NE_Ca scaffolds (Figure 12 a-f). We next quantified the violet fluorescent intensity, which represents the expression of osteogenic markers from immunofluorescence images. A higher mean fluorescence intensity (MFI) values was observed for OPN and BMP-2 in hFob cultured on Coll_pNE_Ca and Coll_pDA_Ca mats, respectively (Figure S4a-c). However, in the case of OCN, a higher MFI was observed for cells seeded on Coll_DA_Ca mats whereas lower MFI values were obtained for cells seeded on Coll_pDA/pNE mats. Western immunoblotting was used to further analyze the expression of OPN by cells seeded on various mats. As shown in Figure 12g, cells seeded on Coll_NE_Ca or Coll_pNE_Ca mats displayed about 1.7 fold higher expression of OPN than cells seeded on TCP or ES_Coll, whereas no substantial difference the OPN expression was observed for cells seeded on Coll_DA_Ca or Coll_pDA_Ca. Together with immunostaining, these results demonstrate excellent osteoconductive properties of norepinephrine-Ca2+ composite structures. Cellular morphology of the hFob seeded on various scaffolds were visualized (11 days p.s.) by FE-SEM (Figure 13 a-f). The results indicated enhanced cell attachment and cell spreading on various collagen scaffolds with the formation of mineral particles. The energy dispersive spectrum of the minerals deposited over the scaffolds by hFob further confirms the formation of calcium phosphate (Figure S5). The morphology of cells seeded on Coll_pDA_Ca and Coll_pNE_Ca scaffolds indicated the presence of more consolidated several layers of fully extended cells interconnected by lamellipodia and secretion of ECM (Figure 13 e, f). An abundant mass of fibers beneath the cells indicated the deposition of ECM (Figure 13f inset), confirming successful cell adhesion and mineralization. Designing scaffolds for bone tissue engineering is complex since the natural composite display gradient mechanical properties and porosity (i.e., cortical bone has higher Young’s modulus and compressive strength than cancellous bone). In addition to biocompatibility, porosity and bioresorbability, ideal scaffolds for bone tissue engineering must match the mechanical properties of the bone tissues.53 Empirically, Engler et al, have shown that matrix with higher stiffness has great potential for osteoblasts differentiation.54 In agreement with their observations, it is possible that the higher Young’s modulus observed for the mineralized scaffolds (Coll_pDA/pNE_Ca) may be responsible for increased hFob adhesion, proliferation and differentiation. The overall obtained results suggested that the mineralized scaffolds promote better biochemical cues for cellular adhesion, proliferation,

32

ACCEPTED MANUSCRIPT mineralization and enhanced expression of osteo-specific proteins proved the potential applications in bone tissue regeneration.

a)

b)

c)

d)

e)

f)

Figure 13. Morphology of hFob seeded on various collagen mats. Representative SEM images showing the features of hFOb after 11 days p.s. (a) TCP, (b) ES_Coll, (c) Coll_DA_Ca, (d) Coll_NE_Ca, (e) Coll_pDA_Ca and (f) Coll_pNE_Ca. Scale bar = 10 µm. 4. CONCLUSIONS Mimicking the formation of vertebrate bone during early stages, we developed a mineralized nanofibrous collagen scaffolds for guided bone regeneration by taking advantage of comproportionation of catecholamines and mineral induction by ammonium carbonate diffusion method (ADM). Morphological analysis and surface chemical characterization confirmed the successful formation of collagen-polycatecholamines-CaCO3 composite structures. This methodology resulted in significant enhancement in the mechanical

33

ACCEPTED MANUSCRIPT properties of collagen without affecting the surface wettability of the composite fibres. Incorporation of polycatecholamines conferred interesting blue and green fluorescent properties enabling the mats with multifunctional characteristics. Biological studies with hFob showed increased cell proliferation and osteogenic differentiation in the mineralized composite mats (Coll_pDA/pNE_Ca) than TCP or pristine collagen mats. Together, the results demonstrate that polycatecholamines-CaCO3 composite collagen nanofibrous mats would provide multifunctional scaffolds properties for possible bone tissue engineering applications. ACKNOWLEDGMENT The authors thank the Translational and Clinical Research Flagship Program of the Singapore National Research Foundation (NMRC/TCR/008-SERI/2013) and administered by the National Medical Research Council of the Singapore Ministry of Health. This work was supported by Co-operative Basic Research Grant from the Singapore National Medical Research Council awarded to RL (NMRC/CBRG/0048/2013). NKV acknowledges funding support from Lee Kong Chian School of Medicine, Nanyang Technological University StartUp Grants (L0412130 & L0412290) and the Ministry of Education Singapore AcRF-Tier I Grant (2014-T1-001-141). DISCLOSURE STATMENT The authors declare no competing financial interests. APPENDIX A. SUPPLEMENTARY DATA Supplementary Information. Figures S1-S5 and Tables S1 –S5. REFERENCES [1]Borgström, F.; Lekander, I.; Ivergård, M.; Svedbom, A.; Bianchi, M. L.; Clark, P.; Cluriel, M. D.; Dimai, H. P.; Jurisson, M.; Kallikorm, R.; Lesnyak, O.; McCloskey, E.; Sanders, K. M.; Silverman, S.; Tamulaitiene, M.; Thomas, T.; Tosteson, A. N.; Jonsson, B.; Kanis, J. A. The International Costs and Utilities Related to Osteoporotic Fractures Study (ICUROS)—quality of life during the first 4 months after fracture. Osteoporos. Int. 2013, 24, 811-823. [2] Ensrud, K. E. Epidemiology of fracture risk with advancing age. J. Gerontol. A Biol. Sci. Med. Sci. 2013, 68, 1236-1242.

34

ACCEPTED MANUSCRIPT [3] Burge, R.; Dawson-Hughes, B.; Solomon, D. H.; Wong, J. B.; King, A.; Tosteson, A. Incidence and economic burden of osteoporosis-related fractures in the United States, 2005-2025. J. Bone Miner. Res. 2007, 22, 465–475. [4] Gao, C.; Wei, D.; Yang, H.; Chen, T.; Yang, L. Nanotechnology for treating osteoporotic vertebral fractures. Int. J. Nanomed. 2015, 10, 5139-5157. [5] Puppi, F.; Chiellini, F.; Piras, A. M.; Chiellini, E. Polymeric materials for bone and cartilage repair. Prog. Polym. Sci. 2010, 35, 403-440. [6] Bose, S.; Roy, M.; Bandyopadhyay, A. Recent advances in bone tissue engineering scaffolds. Trends Biotechnol. 2012, 30, 546-554. [7] Beachley, V.; Wen, X. Polymer nanofibrous structures: Fabrication, biofunctionalization, and cell interactions. Prog. Polym. Sci. 2010, 35, 868-892. [8] Wade, R. J.; Burdick, J. A. Advances in nanofibrous scaffolds for biomedical applications: From electrospinning to self-assembly. Nano Today 2014, 9, 722-742. [9] Araujo, J. V.; Carvalho, P. P.; Best, S. M. Electrospinning of Bioinspired Polymer Scaffolds. Adv. Exp. Med. Biol. 2015, 881, 33-53. [10] Ribba, L.; Parisi, M.; D'Accorso, N. B.; Goyanes, S. Electrospun nanofibrous mats: from vascular repair to osteointegration. J. Biomed. Nanotechnol. 2014, 10, 3508-3535. [11] Kim, H. W.; Lee, H. H.; Knowles, J. C. Electrospinning biomedical nanocomposite fibers of hydroxyapatite/poly(lactic acid) for bone regeneration. J. Biomed. Mater. Res. A 2006, 79, 643–649. [12] Fujihara, K.; Kotaki, M.; Ramakrishna, S. Guided bone regeneration membrane made of polycaprolactone/calcium carbonate composite nano-fibers. Biomaterials 2005, 26, 4139– 4147. [13] Song, J. H.; Kim, H. E.; Kim, H. W. Electrospun fibrous web of collagen-apatite precipitated nanocomposite for bone regeneration. J. Mater. Sci. Mater. Med. 2008, 19, 2925–2932. [14] Zhang, Y.; Venugopal, J. R.; El-Turki, A.; Ramakrishna, S.; Su, B.; Lim C. T. Electrospun biomimetic nanocomposite nanofibers of hydroxyapatite/chitosan for bone tissue engineering. Biomaterials 2008, 29, 4314–4322. [15] Yeo, M. G.; Kim, G. H. Preparation and characterization of 3D composite scaffolds based on rapid-prototyped PCL/-TCP struts and electrospun PCL coated with collagen and HA for bone regeneration. Chem. Mater. 2012, 24, 910-913.

35

ACCEPTED MANUSCRIPT [16] Song, J. H.; Yoon, B. H.; Kim, H. E.; Kim H. W. Bioactive and degradable hybridized nanofibers of gelatin–siloxane for bone regeneration. J. Biomed. Mater. Res. A. 2008, 84,875-884. [17] Gao, C.; Gao, Q.; Li, Y.; Rahaman, M. N.; Teramoto, A.; Abe, K. In vitro evaluation of electrospun gelatin-bioactive glass hybrid scaffolds for bone regeneration. J. Appl. Polym. Sci. 2013, 127, 2588–2599. [18] Kim, H. W.; Song, J. H.; Kim, H. E. Nanofiber generation of gelatin-hydroxyapatite biomimetics for guided tissue regeneration. Adv. Funct. Mater. 2005, 15, 1988–1994. [19] Frohbergh, M. E.; Katsman, A.; Botta, G. P.; Lazarovici, P.; Schauer, C. L.; Wegst, U. G.; Lelkes, P. I. Electrospun hydroxyapatite-containing chitosan nanofibers crosslinked with genipin for bone tissue engineering. Biomaterials 2012, 33, 9167-9178. [20] Gupta, D.; Venugopal, J.; Mitra, S.; Giri Dev, V. R.; Ramakrishna, S. Nanostructured biocomposite substrates by electrospinning and electrospraying for the mineralization of osteoblasts. Biomaterials 2009, 30, 2085-2094. [21] Seyedjafari, E.; Soleimani, M.; Ghaemi, N.; Shabani, I. Nanohydroxyapatite-Coated Electrospun Poly(L-lactide) Nanofibers Enhance Osteogenic Differentiation of Stem Cells and Induce Ectopic Bone Formation. Biomacromolecules 2010, 11, 3118-3125. [22] Prabhakaran, M. P.; Venugopal, J.; Ramakrishna, S. Electrospun nanostructured scaffolds for bone tissue engineering. Acta Biomater. 2009, 5, 2884-2893. [23] Wu, X.; Miao, L.; Yao, Y.; Wu, W.; Liu, Y.; Chen, X.; Sun, W. Electrospun fibrous scaffolds combined with nanoscale hydroxyapatite induce osteogenic differentiation of human periodontal ligament cells. Int. J. Nanomed. 2014, 9, 4135-4143. [24] Ramasamy, J. G.; Akkus, O. Local variations in the micromechanical properties of mouse femur: The involve-ment of collagen fiber orientation and mineralization. J. Biomech. 2007, 40, 910–918. [25] Isaksson, H.; Harjula, T.; Koistinen, A.; Iivarinen, J.; Seppänen, K.; Arokoski, J. P.; Brama, P. A.; Jurvelin, J. S.; Helminen, H. J. Collagen and mineral deposition in rabbit cortical bone during maturation and growth: effects on tissue properties. J. Orthop. Res. 2010, 28, 1626-1633. [26] Lee, H.; Dellatore, S. M.; Miller, W. M.; Messersmith, P. B. Mussel-inspired surface chemistry for multifunctional coatings. Science 2007, 318, 426-430. [27] Kang, S. M.; Rho, J.; Choi, I. S.; Messersmith, P. B.; Lee, H. Norepinephrine: materialindependent, multifunctional surface modification reagent. J. Am. Chem. Soc. 2009, 131, 13224-13225. 36

ACCEPTED MANUSCRIPT [28] Hong, S.; Kim, J.; Na Y. S.; Park, J.; Kim, S.; Singha, K.; Im, G.; Han D.; Kim, W. J.; Lee, H. Poly(norepinephrine): ultrasmooth material-independent surface chemistry and nanodepot for nitric oxide. Angew. Chem. Int. Ed. 2013, 52, 9187-9191. [29] Vidal, C. M.; Aguiar, T. R.; Phansalkar, R.; McAlpine, J. B.; Napolitano, J. G.; Chen, S. N.; Araújo, L. S.; Pauli, G. F.; Bedran-Russo, A. Galloyl moieties enhance the dentin biomodification potential of plant-derived catechins. Acta Biomater. 2014, 10, 3288-3294. [30] He, L.; Mu, C.; Shi, J.; Zhang, Q.; Shi, B.; Lin, W. Modification of collagen with a natural cross-linker, procyanidin. Int. J. Biol. Macromol. 2011, 48, 354-359. [31] Zeugolis, D. I.; Khew, S. T.; Yew, E. S.; Ekaputra, A. K.; Tong, Y. W.; Yung, L. Y.; Hutmacher, D. W.; Sheppard, C.; Raghunath, M. Electro-spinning of pure collagen nanofibres - just an expensive way to make gelatin? Biomaterials 2008, 29, 2293-2305. [32] Matthews, J. A.; Wnek, G. E.; Simpson, D. G.; Bowlin, G. L. Electrospinning of collagen nanofibers. Biomacromolecules 2002, 3, 232-238. [33] Chai, D.; Xie, Z.; Wang, Y.; Liu. L.; Yum Y. Molecular Dynamics Investigation of the Adhesion Mechanism Acting between Dopamine and the Surface of Dopamine-Processed Aramid Fibers. ACS Appl. Mater. Interfaces 2014, 6, 17974-17984. [34] Robinson, D. L.; Venton, B. J.; Heien, Michael L. A. V.; Wightman, R. M. Detecting Subsecond Dopamine Release with Fast-Scan Cyclic Voltammetry in Vivo, Clin. Chem. 2003, 49, 1763-1773. [35] Li, Y.; Tan, Y.; Deng, W.; Xie, Q.; Zhang, Y.; Chen, J.; Yao, S. Electropolymerization of catecholamines after laccase-catalyzed preoxidation to efficiently immobilize glucose oxidase for sensitive amperometric biosensing. Sens. Actuat. B-Chem. 2010, 151, 30-38. [36] Zangmeister, R. A.; Morris, T. A.; Tarlov, M. J. Characterization of polydopamine thin films deposited at short times by autoxidation of dopamine. Langmuir 2013, 29, 86198628. [37] He, S.; Zhou, P.; Wang, L.; Xiong, X.; Zhang, Y.; Deng, Y.; Wei, S. Antibiotic-decorated titanium with enhanced antibacterial activity through adhesive polydopamine for dental/bone implant. J. R. Soc. Interface 2014, 11, 20140169. [38] Hu, X.; Joshi, P.; Mukhopadhyay, S. M.; Higgins, S. R. X-ray photoelectron spectroscopic studies of dolomite surfaces exposed to undersaturated

and super saturated aqueous

solutions. Geochim. Cosmochim. Acta 2006, 70, 3342-3350. [39] Demri, B.; Muster, D. XPS study of some calcium compounds. J. Mater. Process. Technol. 1995, 55, 311-314.

37

ACCEPTED MANUSCRIPT [40] Baltrusaitis, J.; Usher, C. R.; Grassian, V. H. Reactions of sulfur dioxide on calcium carbonate single crystal and particle surfaces at the adsorbed water carbonate interface. Phys. Chem. Chem. Phys. 2007, 9, 3011-3024. [41] Blanchard Jr, D. L.; Baer, D. R. The interaction of Co, Mn and water with calcite surfaces. Surf. Sci. 1992, 276, 27-39. [42] Kouhi, M.; Morshed, M.; Varshosaz, J.; Fathi, M. H. Poly (e-caprolactone) incorporated bioactive glass nanoparticles and simvastatin nanocomposite nanofibers: Preparation, characterization and in vitro drug release for bone regeneration Applications. Chem. Eng. J. 2013, 228, 1057-1065. [43] Chu, D. H.; Vinoba, M.; Bhagiyalakshmi, M.; Baek, I. H.; Nam, S. C.; Yoon, Y.; Kim, S. H.; Jeong, S. K. CO2 mineralization into different polymorphs of CaCO3 using an aueousCO2 system. RSC Adv. 2013, 3, 21722-21729. [44] Kharaziha, M.; Fathi, M. H.; Edris, H. Development of novel aligned nanofibrous composite membranes for guided bone regeneration J. Mech. Behav. Biomed. Mater. 2013, 24, 9-20. [45] Wutticharoenmongkol, P.; Sanchavanakit, N.; Pavasant, P.; Supaphol, P. Preparation and characterization of novel bone scaffolds based on electrospun polycaprolactone fibers filled with nanoparticles. Macromol. Biosci. 2006, 6, 70-77. [46] Tian, M.; Gao, Y.; Liu, Y.; Liao, Y.; Xu, R.; Hedin, N. E.; Fong, H. Bis-GMA/TEGDMA Dental Composites Reinforced with Electrospun Nylon 6 Nanocomposite Nanofibers Containing Highly Aligned Fibrillar Silicate Single Crystals. Polymer (Guildf) 2007, 48, 2720-2728. [47] Zhao, C.; Zuo, F.; Liao, Z.; Qin, Z.; Du, S.; Zhao, Z. Mussel-inspired one-pot synthesis of a fluorescent and water-soluble polydopamine-polyethyleneimine copolymer. Macromol. Rapid Commun. 2015, 36, 909-915. [48] Yang, J.; Zhang, Y.; Gautam, S.; Liu, L.; Dey, J.; Chen, W.; Mason, R. P.; Serrano, C. A.; Schug, K. A.; Tang, L. Development of aliphatic biodegradable photoluminescent polymers. Proc. Natl. Acad. Sci. U S A. 2009, 106, 10086-10091. [49] Xie, Z.; Zhang, Y.; Liu, L.; Weng, H.; Mason, R. P; Tang, L.; Nguyen, K. T.; Hsieh, J. T.; Yang, J. Development of intrinsically photoluminescent and photostable polylactones. Adv. Mater. 2014, 26, 4491-4496. [50] Li, X.; Li, Y.; Chen, X.; Li, B.; Gao, B.; Ren, Z.; Han, G.; Mao, C. Optically monitoring mineralization and demineralization on photoluminescent bioactive nanofibers. Langmuir. 2016, 32, 3226-33. 38

ACCEPTED MANUSCRIPT [51] Kulterer, B.; Friedl, G.; Jandrositz, A.; Sanchez-Cabo, F.; Prokesch, A.; Paar, C.; Scheideler, M.; Windhager, R.; Preisegger, K.H.; Trajanoski, Z. Gene expression profiling of human mesenchymal stem cells derived from bone marrow during expansion and osteoblast differentiation. BMC Genomics 2007, 8, 70. [52] Anderson, H. C.; Sipe, J. B.; Hessle, L.; Dhanyamraju, R.; Atti, E.; Camacho, N. P.; Millan, J. L. Impaired calcification around matrix vesicles of growth plate and bone in alkaline phosphatase-deficient mice. Am. J. Pathol. 2004, 164, 841-847. [53] Bose, S.; Roy, M.; Bandyopadhyay, A. Recent advances in bone tissue engineering scaffolds. Trends Biotechnol. 2012, 30, 546-554. [54] Engler, A. J.; Sen, S.; Sweeney, H. L.; Discher, D. E. Matrix elasticity directs stem cell lineage specification. Cell. 2006, 126, 677-689.

39