Hyaluronic-Acid–Hydroxyapatite Colloidal Gels ... - ACS Publications

Nov 23, 2016 - Native ECM biomaterials are a key protagonist in the area of bone ... We hypothesized that native ECM microparticles could be combined ...
0 downloads 0 Views 3MB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Hyaluronic Acid-Hydroxyapatite Colloidal Gels Combined with Micronized Native ECM as Potential Bone Defect Fillers Stephen Connor Dennis, Jonathan Whitlow, Michael S. Detamore, Sarah L. Kieweg, and Cory Berkland Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b03529 • Publication Date (Web): 23 Nov 2016 Downloaded from http://pubs.acs.org on December 1, 2016

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.

Langmuir 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 47

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

Langmuir

1

Hyaluronic Acid-Hydroxyapatite Colloidal Gels Combined with Micronized Native ECM as Potential Bone Defect Fillers S. Connor Dennis1, 2, Jonathan Whitlow1, 2, Michael S. Detamore3, Sarah L. Kieweg1, 4, Cory J. Berkland1, 2, 5* 1

Bioengineering Program, 2Chemical and Petroleum Engineering Department, 4Mechanical Engineering Department, and 5Pharmaceutical Chemistry Department, University of Kansas, Lawrence, KS 3 Stephenson School of Biomedical Engineering, University of Oklahoma, Norman OK

KEYWORDS: Hydroxyapatite; Hyaluronic Acid; Colloidal Gel; Native ECM Biomaterials, Demineralized Bone Matrix; Decellularized Cartilage; Herschel-Bulkley Fluid; Tissue Defect Filler Abbreviated Title: Micronized ECM Colloidal Gel Composites as Bone Fillers *Corresponding author: Dr. Cory Berkland Ph.D. 2030 Becker Drive, Lawrence, KS, USA 66047 Email: [email protected] Phone: 785-864-1455 Fax: 785-864-1454

ACS Paragon Plus Environment

Langmuir

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 47

2

ABSTRACT One of the grand challenges in translational regenerative medicine is the surgical placement of biomaterials. In bone regeneration in particular, malleable and injectable colloidal gels are frequently designed to exhibit self-assembling and shear-response behavior which facilitates biomaterial placement in tissue defects. The current study demonstrated that by combining native extracellular matrix (ECM) microparticles, i.e., demineralized bone matrix (DBM) and decellularized cartilage (DCC), with hyaluronic acid (HA) and hydroxyapatite (HAP) nanoparticles, a viscoelastic colloidal gel consisting exclusively of natural materials was achieved. Rheological testing of HA-ECM suspensions and HA-HAP-ECM colloidal gels concluded either equivalent or substantially higher storage moduli (G′ ≈ 100-10,000 Pa), yield stresses (τy ≈ 100-1000 Pa), and viscoelastic recoveries (G′Recovery ≥ 87%) in comparison with controls formulated without ECM, which indicated a previously unexplored synergy in fluid properties between ECM microparticles and HA-HAP colloidal networks. Notable rheological differences were observed between respective DBM and DCC formulations, specifically in HAHAP-DBM mixtures, which displayed a mean three-fold increase in G′ and a mean four-fold increase in τy from corresponding DCC mixtures. An initial in vitro assessment of these potential tissue fillers as substrates for cell growth revealed that all formulations of HA-ECM and HAHAP-ECM showed no signs of cytoxicity and appeared to promote cell viability. Both DBM and DCC colloidal gels represent promising platforms for future studies in bone and cartilage tissue engineering. Overall, the current study identified colloidal gels constructed exclusively of natural materials, with viscoelastic properties that may facilitate surgical placement for a wide variety of therapeutic applications.

ACS Paragon Plus Environment

Page 3 of 47

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

Langmuir

3

1. INTRODUCTION Self-assembling and shear-responsive biomaterials possess favorable rheological and viscoelastic properties for filling bone tissue defects.1 Such materials may be designed to be injectable to facilitate minimally invasive surgery that reduces the risk of infection and scar formation with possible benefits of reducing patient discomfort and treatment costs.1 Furthermore, these malleable biomaterials may be designed to exhibit viscoelastic recovery properties which improve the degree of material retention within a defect site after surgical placement. Although numerous non-setting bone pastes already exist in the orthopedic market including DBX® (MTF/Synthes), Dynagraft II (Integra Orthobiologics), Grafton® Gel (Osteotech), Puros® DBM (Zimd TricOS (Baxter)2, the majority of these products consist of microparticle suspensions in viscous carrier fluids. These microparticle formulations have been reported to exhibit undesirable phase-separation upon injection (e.g., filter-pressing) and after placement in the tissue defect (e.g., sedimentation)3 which can result in poor retention within the defect site.4 In addition, some ceramic microparticles have displayed poor resorption kinetics in vivo, leading to a delay in bone healing.5 A promising strategy to enhance material retention within wound sites is to improve the bulk rheological properties of these colloidal gels and pastes by formulation with nanoparticles rather than microparticles.6,

7

The higher surface energy of nanoparticles in comparison to

microparticles also makes them more chemically reactive and gives nanoparticles a higher affinity to bind and interact with other materials. In a colloidal gel, the bulk rheological properties rely on the cohesive strength of physical crosslinking which depend on electrostatic forces, van der Waals attraction, and steric hindrance within the gel.1, 6, 7 Previous investigations have explored leveraging these interactions with nanoparticles composed of poly(lactic-co-

ACS Paragon Plus Environment

Langmuir

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 47

4

glycolic) acid6, gelatin8, 9, dextran10, and hydroxyapatite (HAP)7,

11

to form cohesive colloidal

gels. Shear-responsive HAP gels are of special interest since this biomolecule closely resembles biological apatite native to bone extracellular matrix (ECM).12 HAP-based nanocomposite gels exhibit desirable mechanical properties for bone tissue engineering and have even shown to form structures with both macroporosity and submicroporosity.13 Our previous results have furthermore demonstrated that the rheology of pure HAP colloidal gels can be modulated with the addition of native glycosaminoglycan (GAG) polymers, such as hyaluronic acid (HA), to elicit varying degrees of viscoelasticity without compromising the material’s inherent yield stress.7 These polymer-particle colloidal gels exhibited synergistic bulk fluid properties likely due to physical crosslinking and bridging flocculation between multivalent GAG polyanions and HAP nanoparticles. Not only does the addition of nanoparticles to colloidal gels allow for precise tuning of the bulk physical properties, the high surface area-to-volume ratio with respect to microparticles can also be exploited to expedite material resorption and to modulate the release of incorporated bioactive molecules.8, 9 While the loading and release of bioactive growth factors has been a major focus of colloidal technology, a lesser explored facet of research is focused on utilizing the osteogenic potential of biomaterials derived from ECM. Native ECM biomaterials are a key protagonist in the area of bone regeneration since they possess both conductive and inductive signaling moieties that promote bone healing and remodeling; the sheer complexity of designing biomimetics that replicate the physical, mechanical, and chemical properties of native tissues has urged many researchers to investigate tissue engineering strategies with raw ECM materials rather than synthetic materials.14-16 Demineralized bone matrix (DBM) from allogeneic sources remains one of the most utilized native ECM biomaterials for bone regeneration2 since it resembles native

ACS Paragon Plus Environment

Page 5 of 47

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

Langmuir

5

bone structure.17, 18 Not only has the osteogenic potential of DBM been correlated to that of bone morphogenetic proteins, but recent evidence has also shown that the structural components of native ECM influence cell signaling and in turn processes such as cell adhesion, differentiation, and migration.2, 16, 19, 20 An alternative and underexplored strategy that could possibly promote or stimulate the natural remodeling functions of fracture healing and thus enhance regeneration in non-load bearing critical sized defects is the use of cartilage-based ECM. Decellularized cartilage (DCC) is one such ECM biomaterial that is rich in both chondroinductive and chondroconductive potential.14-16 While DCC is already used in commercial cartilage regeneration strategies, current research has not yet clarified its potential role in enhancing endochondral ossification in the regeneration of bone defects. The coupling of either DBM or DCC microparticles within a colloidal fluid carrier, such as GAG-HAP, may act to enhance both the physical (e.g., rheological) and chemical (e.g., osteoinductive) properties of traditional bone pastes. Our preceding study analyzed GAG-HAP colloidal gels and selected promising formulations as candidates for fillers of non-load bearing tissue defects.7 The current study expands on the previous, to present a new tissue engineering strategy that is the first attempt to incorporate native ECM into colloidal gel composites. We hypothesized that native ECM microparticles could be combined with HA-HAP mixtures to yield viscoelastic gels with the requisite rheological properties for potential use as bone defect fillers. A primary goal was to elucidate the relationship between bulk fluid properties and relative HA polymer, HAP nanoparticle, and ECM microparticle composition. An initial theory was that the addition of ECM microparticles would disrupt colloidal interactions between HA and HAP, leading to inferior rheological and viscoelastic performance despite the potential increase in osteogenic

ACS Paragon Plus Environment

Langmuir

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 47

6

capacity. The aim of this assessment was to identify biocompatible colloidal ECM composites with the fluid properties desirable for future investigation of this platform in bone and cartilage regeneration applications.

ACS Paragon Plus Environment

Page 7 of 47

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

Langmuir

7

2. EXPERIMENTAL 2.1. Colloidal Materials Hydroxyapatite was purchased as a powder (Davg ≤ 200 nm (BET Analysis); Sigma-Aldrich). Hyaluronic acid at 35 kDa, 350 kDa, and 1500 kDa (Lifecore Biomedical; Chaska, MN) was purchased as a sodium salt.

2.2. Demineralized Bone Matrix and Decellularized Cartilage Procurement Human demineralized bone matrix was procured as a course ground powder (Biomet, Irvine, CA). DBM particles were cryo-ground into a fine powder with a freezer-mill (SPEX, SamplePrep, Metuchen, NJ), and particles were sieved (Spectra/Mesh Woven Filters, Spectrum Laboratories, Inc., Rancho Dominguez, CA) to exclude particles measuring above 350 µm to ensure the consistency of fluid properties and reproducibility of each gel for subsequent rheometer experiments. Decellularized cartilage procured from porcine knee and hip joints was purchased from a local abattoir following sacrifice (120 kg, mixed breed, mixed gender) (Bichelmeyer Meats, Kansas City, KS). Articular cartilage from both the knee and hip joints was carefully removed and collected using scalpels. The cartilage was rinsed in phosphate buffered saline (PBS) and stored at -20°C. Following freezing, the cartilage was coarsely ground using a cryogenic tissue grinder (BioSpec Products, Bartlesville, OK). The coarsely ground tissue was packaged into dialysis tubing (3500 MWCO) packets for decellularization. An adapted version of our previously established cartilage decellularization method was used which included reciprocating osmotic shock, detergent, and enzymatic washes.21 Reagents were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise noted. All steps of decellularization were carried out under agitation (200 rpm) at 21°C unless indicated otherwise.

ACS Paragon Plus Environment

Langmuir

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 47

8

First, the cartilage packets were placed in hypertonic salt solution (HSS) overnight to disrupt membranes and lyse the cells. Following HSS treatment, the tissue was subjected to 2 cycles of reciprocating Triton-X 100 (0.05% v/v) and HSS treatments to further break down cellular membranes. The tissue was then treated with benzonase (0.0625 KU ml-1) overnight at 37°C to fragment any nucleic acid remaining in the tissue. Sodium-lauroyl sarcosine (NLS, 1% v/v) was subsequently used overnight to further solubilize and remove cells. Next, the tissue was washed with 40% ethanol, followed by organic exchange resins to remove all organic solvents. To conclude the decellularization process, the tissue was removed from the dialysis tubing packages and rinsed with deionized water before freezing. After decellularization, the tissue was lyophilized for 48 hours and cryo-ground into a fine powder with a freezer-mill (SPEX SamplePrep, Metuchen, NJ). Particles were sieved (Spectra/Mesh Woven Filters, Spectrum Laboratories, Inc., Rancho Dominguez, CA) to exclude particles measuring above 350 µm. Size measurements of micronized DCC and DBM particles were taken via scanning electron microscopy (Carl Zeiss Leo 1550 Field Emission SEM).

2.3. Preparation of Colloidal Gels HA polymers were combined with HAP nanoparticles in PBS solution (pH = 7.4, 150 mM NaCl) according to our previously published protocols.7 HA concentrations were varied between 0-15%, 0-5%, and 0-2% (w/v) in colloidal mixtures containing 35 kDa, 350 kDa and 1500 kDa HA polymers, respectively. The polymer to nanoparticle (HA:HAP) weight ratio (w:w) was adjusted by the incremental addition of HAP particles to HA solutions. These mixtures were compared to pure component controls (HA and HAP), respectively. The overall volume fraction of HAP (ΦHAP = VHAP/VTOTAL) was calculated from particle density and resulting mixture volume measurements. Additional sizing and zeta-potential measurements of HAP nanoparticles

ACS Paragon Plus Environment

Page 9 of 47

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

Langmuir

9

in the presence of HA polymers at various molecular weights (35, 350, and 1500 kDa) were conducted using dynamic light scattering (Brookhaven; ZetaPALS) to analyze particle-polymer interactions. Lyophilized DBM and DCC microparticles were formulated as HA-ECM suspensions and HA-HAP-ECM mixtures to elucidate the effect of the ECM on rheological and viscoelastic properties (Tables 1 and 2). This was achieved by maintaining constant media volume (1 ml), HA concentration (1-2% w/v) and HAP volume fraction (ΦHAP = 30-32%) in the tested fluids. The increment of ECM incorporation was varied in terms of dry mass fraction (wDCC = 3-9% and wDBM = 3-28%) and in terms of volume fraction (ΦDCC = 15-45% and ΦDBM = 5-45%). ECM volume fraction was estimated from measured water content and density of respective particles following hydration. ECM hydration was assumed to be 100% in all formulations. In reality, the incorporated HA polymers likely exhibit some amount of osmotic pressure on nearby ECM particles, leading to a hydration equilibrium state slightly below 100%. Although the dynamics of this equilibrium were not characterized in the present study, the amount of incorporated media was well above the hydration limit of DCC and DBM. At this limit, it could be assumed that all available media in the formulation would reside within ECM particles, leaving no free media to solubilize the carrier fluid. Homogeneous suspensions and colloid mixtures were prepared by manual stirring (5 min) at ambient conditions and stored at 4°C. Samples were allowed to equilibrate to ambient conditions (2 hr) before testing.

2.4. Swelling Characterization Relative swelling ratios (S) of colloidal gels were determined by placing 1.0 mL of PBS (pH = 7.4, 150 mM NaCl) on top of 0.5 mL of material contained in a 2 mL Eppendorf tube. Tubes were then constantly agitated (24 hr, 100 rpm, 37°C) in an incubator shaker (New Brunswick

ACS Paragon Plus Environment

Langmuir

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 47

10

Scientific; Excella E24). Swelling ratio (S = (M(swollen)-M(before))/M(before)) was determined from the initial (M(before)) and final (M(swollen)) mass of the material as described by Holland et al.22 The final weight was determined after removing excess PBS from tube and drying the surface of gel with evaporative paper.

2.5. Rheological Analysis The rheological and viscoelastic properties of HA-HAP colloidal gel formulations (Table S1) consisting of 35, 350, and 1500 kDa HA were characterized using a controlled stress rheometer (TA Instrument; AR2000) and compared to an HAP colloid control group. In addition, the rheological and viscoelastic performance of HA-ECM microparticle suspensions and HAHAP-ECM colloidal gel formulations (Table S2) was tested and compared to respective HA (1500 kDa), HAP, and HA-HAP controls. All measurements were collected at a gap distance of 500 µm using a roughened stainless steel plate geometry (20 mm diameter) and roughened base plate to minimize slip conditions. All samples were tested at 37°C. Viscoelastic recovery kinetics following temporary disruption of the colloidal gel network were determined by measuring viscoelastic properties as described by Ozbas et al.23 Initially, an oscillatory stress sweep (1-10,000 Pa) was performed at a constant frequency (1 Hz) to determine the storage modulus (G′), loss modulus (G″), loss angle (δ), and yield stress (τy) within the linear viscoelastic (LVE) region of each fluid.24 Within the LVE region, strain measured during small external oscillatory stress is controlled by rates of spontaneous rearrangements or relaxations in the fluid, and thus can be assumed to represent the resting, or quiescent, state of the fluid.24 The upper boundary limit of the LVE region was used to approximate the yield stress, τy

25

, in the material.26 To ensure internal consistency in estimating this boundary limit, τy was

calculated from the external oscillatory stress corresponding to a 10% reduction in G′ averaged

ACS Paragon Plus Environment

Page 11 of 47

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

Langmuir

11

within the fluid’s LVE region. Above the LVE, stress-induced deformations cause temporary disruptions in the fluid microstructure. Subsequent gel recovery time sweeps were conducted as four alternating rounds of intense oscillatory disruption above the LVE region (30 s, 1000 Pa, 1 Hz) and recovery within the LVE region (5 min, 10-25 Pa, 1 Hz) at 37°C following pre-shear (1 min, 100 s-1) and equilibration (5 min). Recovery was assessed 5 minutes after disruption and expressed as a percentage (G′Recovery = G′(Final)/G′(Initial) •100%) compared to the initial G′ of the mixture.

2.6. In Vitro Cell Culture Human umbilical cord mesenchymal stromal cells (hUCMSCs) were harvested from Wharton’s Jelly of a human umbilical cord with informed consent (IRB #15402) as detailed previously by Mellott et al.27 The cord was procured from a male born at full term and delivered by Caesarean section. Harvested cells were cultured for one passage (P1) and then transferred to freezing medium and into cryogenic vials for storage in liquid nitrogen until further use. Prior to experiments, cells were thawed and expanded from P2 to P5 in culture medium consisting of 10% fetal bovine serum and 1% penicillin/streptomycin in low-glucose DMEM (ThermoFisher).

2.7. Cell Seeding and Viability hUCMSCs were seeded onto each colloidal gel formulation (Table S2) at a seeding density of approximately 5 x 104 cells/cm2. Briefly, colloidal gels were deposited into the wells of a nontreated 96-well plate each at a volume of 50 µL (n = 3) for each gel formulation (Table S2). In order to produce a consistently thin gel layer in each well, the 96-well plate was centrifuged at 3000 rpm for 2 minutes. The gels were sterilized under UV light for 10 minutes, and subsequently incubated at room temperature after the addition of 200 µL of PBS with 1%

ACS Paragon Plus Environment

Langmuir

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 47

12

penicillin/streptomycin to each well. Prior to cell seeding, the PBS was carefully aspirated and replaced with 200 µL of complete growth medium. Cells were then cultured in monolayer for 72 hours, at which point the viability of hUCMSCs on each suspension or colloidal gel was assessed by live/dead assay. All wells were incubated with live/dead reagent (2 µM calcein AM, 4 µM ethidium homodimer-1; Molecular Probes) for 15 minutes and analyzed with fluorescence microscopy (Zeiss Axio Observer A1; Carl Zeiss Microscopy, Thornwood, NY).

2.8. Statistical Analysis All measurements were performed in triplicate (n = 3) and depicted as average ± standard deviation (SD) unless stated otherwise. Statistical analyses of data were performed using oneway analysis of variance (ANOVA) and Tukey’s HSD was used post-hoc to compare differences between individual groups. A p-value (p < 0.05) was accepted as statistically significant.

ACS Paragon Plus Environment

Page 13 of 47

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

Langmuir

13

3. RESULTS 3.1. Characterization of ECM Microparticles and HAP Nanoparticles 3.1.1. DCC, DBM, and HAP Sizing SEM was used to verify particle size of DCC, DBM, and HAP (Figure 1A-C). Both DCC and DBM dry powders exhibited polydisperse size (Davg ≈ 10-100 µm) and heterogeneous morphology. All imaged particles appeared to be below the mesh screen size (< 350 µm), which verifies the effectiveness of the size exclusion sieving procedure. HAP nanoparticles displayed spherical morphologies with polydisperse diameters at values within close range of the supplier’s specified value (Davg ≤ 200 nm (BET); Sigma-Aldrich); however, a small fraction of particles were observed to exceed this specification. HA(35 kDa)-HAP was also imaged with SEM (Supplementary Figure S1) which again confirmed the spherical morphology and size of HAP and also showed the aggregation of HAP nanoparticles around HA polymer with the colloidal gel network.

3.1.2. Effective Diameter and Zeta Potential Dynamic light scattering was used to measure HAP particle size (mm) and zeta potential (mV) in the presence of solubilized 35, 350, and 1500 kDa HA. Dilute suspensions (0.167 mg/mL) of pure HAP particles were combined with HA at a 1:1 weight ratio. Both the effective diameter and zeta potential magnitude increased with HA molecular weight (Table 1). However, only the colloidal gels formed with 350 and 1500 kDa HA resulted in a significant increase in measured particle size and zeta potential (p < 0.01) compared to pure HAP suspensions.

3.2. Swelling Characterization

ACS Paragon Plus Environment

Langmuir

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 47

14

Relative swelling of HA-HAP formulations (Figure 2) composed of 35, 350, and 1500 kDa HA increased with a clear dependence on the volume fraction of HAP (ΦHAP). While sedimentation and swelling tolerances were initially set (-20% ≤ S ≤ 20%) in an attempt to identify suitable HA-HAP candidates for placement of material at a bone defect surgical site, the most desirable colloidal mixtures exhibited the least amount of swelling or sedimentation (S ≈ 0%). Swelling increased with increasing HA molecular weight but could be maintained within the swelling tolerances by decreasing polymer concentration. HA-HAP formulations at all tested molecular weights demonstrated minimized swelling and sedimentation at ΦHAP ≈ 0.30. Therefore, this value was targeted in subsequent rheological and viscoelastic characterization studies. Pure DCC and DBM water content and density upon hydration was also measured following the swelling protocol. Water content in hydrated particles was determined to be 80.5 ± 1.0% (w/w) and 39.5 ± 2.1% for DCC and DBM, respectively. The corresponding densities of hydrated DCC and DBM were 1.09 and 1.06 g/mL.

3.3. Rheological and Viscoelastic Characterization 3.3.1. HA Molecular Weight Studies 3.3.1.1. Storage Modulus (G′) and Loss Angle (δ) Measuring shear strain in response to an increasing external oscillatory shear stress allowed for quantification of relevant viscoelastic properties, including G′, G″, δ, and τy, within and above the LVE region of tested HA-HAP fluids with varying HA molecular weight (Table S1). The presence of either HA or HAP resulted in measurable G′ and G″, indicating that all tested fluids resided within the viscoelastic spectrum (Figure 3). A significant increase in G′ was observed when HAP was added to pure HA solutions, regardless of polymer molecular weight, to form

ACS Paragon Plus Environment

Page 15 of 47

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

Langmuir

15

HA-HAP colloidal gels. Furthermore, there was a significant increase (p < 0.01) in G′ for HA(350)-HAP and HA(1500)-HAP colloids compared to the pure HAP control fluid. This change in G′ for these colloids appeared to be combinatorial increase, such that G′HA-HAP >> G′HA + G′HAP. The measured δ in HA-HAP colloids was 18°, 12°, and 13° for HA 35, 350, and 1500 kDa formulations, respectively. These values were more similar to the loss angle of pure HAP (δHAP = 7°) than respective pure HA solutions (δHA35 = 89°, δHA350 = 78° and δHA1500 = 37°). 3.3.1.2. Yield Stress (τy) Yield stress (Figure 3) of each tested fluid was approximated by the upper boundary limit of the LVE region obtained from the stress sweep profile. Measureable τy amongst pure HA solutions was only observed in 1500 kDa HA at 1% w/v. All tested formulations with HAP displayed a measurable yield stress. In the 1500 kDa HA-HAP colloid where both pure HA and pure HAP controls possessed independent τy values, the resulting τy appeared to be an additive property of the fluid, such that τy (HA-HAP) ≈ τy (HA) + τy (HAP).

3.3.1.3 Viscoelastic Recovery (G′Recovery and δRecovery) Repeated oscillatory disruption of the fluid microstructure did not affect the ability of these colloids to recover their resting viscoelastic properties (Figure 4A). All tested mixtures exhibited nearly complete recovery of G’ (≥ 98%) within 5 minutes (Figure 4B) after intense oscillatory disruption. Each formulations displayed a shift toward more viscous behavior during disruption (δDisruption  90°), and likewise exhibited a shift toward more elastic behavior during recovery (δRecovery  0°) (Figure 4C). 3.3.2. HA-ECM and HA-HAP-ECM Viscoelastic Studies 3.3.2.1. Storage Modulus (G′) and Loss Angle (δ)

ACS Paragon Plus Environment

Langmuir

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 47

16

Based on the rheological results of HA-HAP colloidal gels, HA molecular weight of 1500 kDa was chosen for subsequent use in HA-ECM and HA-HAP-ECM studies (Table S2). A significant increase in G′ was observed (p < 0.01) upon addition of DCC and DBM in HA-ECM microparticle suspensions (Table 2). G′ increased with ECM concentration in the suspension. While there was no significant difference between HA-DCC and HA-DBM formulations with respect to ECM mass fraction, HA-DBM exhibited a significantly higher G′ than corresponding HA-DCC (p < 0.05) at ΦECM = 0.30 and 0.45. On the other hand, no trends based on ECM concentration were observed in G′ of HA-HAP-ECM mixtures (Table 3). However, all HAHAP-DBM formulations exhibited significantly higher G′ values than both HAP (p < 0.01) and HA-HAP (p < 0.05) controls. HA-HAP-DCC mixtures did not have significantly different G′ values compared to HA-HAP, but they did exhibit significantly higher G′ than the HAP (p < 0.05) control. Above ΦDBM = 0.30, HA-HAP-DBM appeared to incompletely hydrate during reconstitution and fractured upon mixing. Consequently, these samples could not be tested on the rheometer. While the addition of DCC particles to HA solutions and HA-HAP colloids did not significantly change δ values (Supplementary Figure S2), DBM microparticles exhibited significantly more elastic behavior than respective HA (p < 0.01) control. DBM formulations also resulted in more elastic behavior than corresponding DCC microparticles in HA (p < 0.01) and HA-HAP colloids (p < 0.01) at a given mass and volume fraction. However, both HAP and HA-HAP were observed to have similar elastic behavior to HA-HAP-DBM formulations.

3.3.2.2. Yield Stress (τy) While measureable τy was observed in all formulations tested (Tables 2 and 3), no significant differences were observed when DCC or DBM was added to HA, except at ΦDBM =

ACS Paragon Plus Environment

Page 17 of 47

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

Langmuir

17

0.45 (p < 0.01). HA-HAP-DCC mixtures only showed a significant increase in τy at the highest tested volume fraction of DCC (ΦDCC = 0.45) compared to HA, HAP, and HA-HAP controls (p < 0.01) (Table 3). Furthermore, increasing DCC concentration in the HA-HAP colloid appeared to result in an increased τy. Interestingly, the opposite trend occurred for HA-HAP-DBM mixtures, in which an increasing DBM concentration resulted in a decreased τy. Nonetheless, all tested HAHAP-DBM mixtures exhibited significantly higher τy than HAP and HA-HAP controls (p < 0.01).

3.3.2.3. Viscoelastic Recovery (G′Recovery and δRecovery) Similar to HA-HAP molecular weight studies, repeated disruption of HA-ECM and HAHAP-ECM did not affect the ability of these fluids to recover their resting viscoelastic properties (Figure 5A). All HA-ECM and HA-HAP-ECM fluids and corresponding controls exhibited nearly complete recovery of G′ (≥ 87%) within 5 minutes (Figure 5B). Recovery experiments with DCC and DBM were only run at an equivalent volume fractions (ΦECM = 0.15) since few differences were observed in previous ECM viscoelastic stress sweep profiles. All formulations displayed a shift toward more viscous behavior during disruption (δDisruption  90°), and likewise exhibited a shift toward more elastic behavior during recovery (δRecovery  0°) (Figure 5C). 3.3.2.4. Needle and Syringe Fluid Extrusion Study Each formulation from the viscoelastic recovery study above was subsequently tested for injectability. All fluids were successfully loaded into 1 mL syringes and readily extruded through attached 18-gauge needles (Figure 6A). Qualitatively, only HA-HAP and HA-HAP-ECM mixtures appeared to exhibit substantial shape retention following extrusion, while HA, HAP, and HA-ECM suspensions appeared to exhibit primarily viscous behavior (Figure 6A).

ACS Paragon Plus Environment

Langmuir

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 47

18

Furthermore, only HA-HAP and HA-HAP-ECM could be subsequently molded into shapes following extrusion (Figure 6B) and maintain shape in ambient conditions following handling (Figure 6C). 3.4 In Vitro Cell Viability Assay In vitro biocompatibility of the gels was assessed by a live/dead assay 72 hours after seeding human umbilical cord mesenchymal stromal cells (hUCMSCs) onto each HA, HAP, HA-HAP, HA-ECM, and HA-HAP-ECM formulation (Table S2). The live/dead assay enabled the detection of viable cells by green fluorescence versus dead cells by red fluorescence. The HAECM and HA-HAP-ECM groups, even those of the highest ECM concentration, demonstrated negligible cytotoxicity, as very few dead cells were observed (Figure 7). Additionally, the HA and HAP groups also displayed minimal cell death, further indicating the minimal cytotoxicity of the individual components of the colloidal gels. Autofluorescence of the materials, primarily HAP, created background interference that prevented image analysis to accurately quantify viable versus dead cells in each sample. While numerical trends could not be calculated, qualitative observations of live/dead images corresponding to each formulation still provided overall confirmation of biocompatibility (Supplementary Figure S3-S4). Several formulations that were included in this experiment, for example the HAP group, were suspensions rather than colloidal gels. In the case of HAP, due to the lack of a colloidal network to support cell growth, the cells grew directly on the surface of the polystyrene wells, and the HAP remained in suspension along with the cell culture media, however; cells were still in direct contact with the biomaterials during the entire time period of the live/dead assay.

4. DISCUSSION

ACS Paragon Plus Environment

Page 19 of 47

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

Langmuir

19

We previously investigated the bulk rheological properties of GAG-HAP colloidal gels with a particular focus on identifying the leading gel formulations with fluid properties desirable for application as bone defect fillers.7 A focal point in the design of GAG-HAP tissue defect fillers was to formulate the colloidal gels with viscoelastic recovery, such that the gels could be injected (i.e., for surgical placement into tissue defects) and rapidly recover their original bulk fluid properties. The next step in advancing GAG-HAP gels towards use as a tissue engineering platform is to first assess the change in bulk and microstructure fluid properties of the gels upon the addition of bioactive molecules. The current study explores the formulation of HA-HAP colloidal gels with osteogenic ECM microparticles, and to the best of our knowledge, represents the first reported attempt to incorporate native ECM microparticles into colloidal gel composites. We initially hypothesized that native ECM microparticles could be introduced into HA-HAP mixtures to form colloidal gels with the desired rheological properties for application as bone defect fillers. However, based upon our observations of the microscale particle interactions in GAG-HAP gels that we previously reported, we theorized that the addition of ECM microparticles to HA-HAP colloidal formulations could disrupt physical crosslinking mechanisms between HA and HAP, possibly resulting in inferior bulk fluid properties, indicated by decreased G’ and τy values, in comparison with HA-HAP controls formed without ECM. In contrast to the expected results based on this theory, the addition of ECM instead improved the viscoelastic properties of all formulations with respect to control fluids formulated without ECM; HA-HAP-ECM and HA-ECM formulations exhibited equivalent or higher G’, τy, and viscoelastic recovery (Tables 2 and 3). Several possible explanations for these unexpected fluid behaviors can be drawn by first reevaluating the conclusions of our previous GAG-HAP study.7

ACS Paragon Plus Environment

Langmuir

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 47

20

The development of potential osteogenic tissue fillers with GAG polymers and HAP nanoparticles primarily relies upon the non-covalent association of HA polymer and HAP nanoparticles to form polymeric colloidal networks. The effects of HA polymer molecular weight on solution rheology and viscoelasticity have been thoroughly characterized in previous studies.28,

29

At non-dilute concentrations, HA solutions exhibit complex, non-Newtonian

behavior due to increased intermolecular entanglements and self-associations of polymer chains in the solution.28, 29 Highly viscous and shear-thinning HA solutions of varying molecular weight and concentration have been studied for biomedical applications such as synovial fluid supplements and as inert carrier fluids in tissue engineered scaffolds.2,

30

More recently, our

group studied the effects of HA (35 kDa) in HAP colloidal gels for tissue defect filling.7 As established in our previous analyses of GAG-HAP gels, rheological and light scattering data supported the hypothesis that ‘bridging’ flocculations formed by the absorption of the multivalent HA polymers to surrounding HAP nanoparticles were largely responsible for the formation of the polymer-particle colloidal network. The interaction between GAG polymers and HAP nanoparticles led to a combinatorial increase in corresponding bulk fluid rheology, which was mapped across an extensive array of concentrations and particle-polymer ratios to fit potential tissue filling parameters.7 Although many recently reported strategies for creating tissue filling nanocomposites require synthetic polymers and nanofillers to achieve the desired mechanical properties, as reviewed in detail by Peng et al.,31 our recent investigation concluded that physical crosslinking between natural GAG polymers and HAP was suitable for creating a range of potential bone defect fillers.7 Initial light scattering experiments in the current study showed that similar polymer-particle colloidal interactions existed within higher molecular weight HA (350 and 1500 kDa)

ACS Paragon Plus Environment

Page 21 of 47

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

Langmuir

21

formulations. The inclusion of HA (350 and 1500 kDa) at dilute HAP concentrations significantly increased the overall effective diameter and zeta potential (p < 0.01) of HAP particles (Table 2) compared to HA 35 kDa and pure HAP control. While increased zeta potential at dilute concentrations may have corresponded to the increased attractive strength of bridging flocculation at high particle concentrations, the resultant increase in particle diameter may have acted to inhibit particle flocculation via steric repulsion32,

33

and thus compromised

beneficial particle-polymer interactions seen with higher molecular weight HA. While strengthening inter-particle attraction would hypothetically lead to increased elastic behavior in tested viscoelastic colloidal fluids, characterized by a higher G’ and τy along with lower δ, increasing steric repulsion between particles would likely lead to an opposing viscous shift. Swelling studies (Figure 2) revealed that HA-HAP formulated with lower concentrations of high molecular weight HA (350 and 1500 kDa) exhibited similar spanning flocculation networks as HA-HAP formulated with 35 kDa HA polymer. Furthermore, corresponding viscoelastic studies (Figures 3 and 4) showed that the colloidal gels formulated with higher molecular weight HA (350 or 1500 kDa) exhibited an increased G’ by two orders of magnitude from the G’ of the colloidal gel formed with 35 kDa HA. For instance, the G’ of HA(35 kDa)-HAP was 2.79 x 103 Pa, while the G’ of HA(350 kDa) was 3.133 x 105 Pa. Collectively, this evidence supports the expected result that higher molecular weight HA strengthens the resulting bridging flocculations in HA-HAP colloidal gels. While some steric repulsive effects may have been present, their contribution to the resulting bulk rheological properties may have been outweighed by attractive polymer-particle interactions. Ultimately, these molecular weight studies concluded that equivalent rheological properties of HA-HAP colloids formed with low concentrations of high molecular weight HA could be

ACS Paragon Plus Environment

Langmuir

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 47

22

matched with high concentrations of low molecular weight HA. Instead of using 35 kDa HA, lesser amounts of 350 and 1500 kDa HA were used to achieve the rheological performance relevant for surgical placement of these potential tissue fillers. The advantages of using 350 and 1500 kDa HA compared to lower molecular weight formulations were (i) the potential to achieve wider ranges of rheological properties for surgical applications and (ii) less material required, which lead to smaller increases in osmolality of the colloid due to the addition of HA polyelectrolytes and associated sodium ions in solution. To apply these two advantages to their fullest extent, the highest molecular weight of HA, 1500 kDa, was chosen for the remainder of the current study for all experiments with ECM microparticles and HAP colloids. Micronized DBM particles have been used extensively in bone tissue engineering applications.2 While DBM has often been formulated with inert polymer carrier fluids such as HA and glycerol, combining the material with colloidal fluids has to date remained underexplored. Microparticles suspended in viscous media do not remain in suspension like colloidal nanoparticles undergoing Brownian motion,1,

34

resulting in time dependent phase

separation and heterogeneity. As a result, microparticle suspensions must be formulated near or above a critical concentration, the percolation threshold (Φp), such that particles are suppressed by the crowding or caging effect of neighboring particles and exhibit a randomly packed continuous network that spans the sample volume.35 While pure spheres have a geometric Φp threshold around 0.29, ellipsoids of various aspect ratios have lower thresholds. SEM images of micronized DCC and DBM (Figure 1) revealed slight ellipsoid morphology that could be approximated as having aspect ratios between 1/3 and 3. The corresponding Φp values for these morphologies were reported to be around 0.22.35

ACS Paragon Plus Environment

Page 23 of 47

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

Langmuir

23

HA-ECM suspensions and HA-HAP-ECM mixtures were formulated to study the effect of ECM, near ΦP, on rheological and viscoelastic properties of the fluids during rest and during microstructure disruption. Since DCC and DBM exhibit different water contents and densities upon hydration, the effect of ECM on bulk fluid properties was directly compared in terms of both mass and volume fraction (Supplementary Tables 1 and 2). As mentioned above, based on the rheological behaviors observed in our previous GAG-HAP study, we theorized in the current study that the addition of ECM microparticles could effectively disrupt the continuity of the respective carrier fluid because shear stress experienced locally by a polymer that is confined between particles could be much larger than the bulk external shear stress,24, 36 leading to inferior viscoelastic properties compared to HA, HAP, and HA-HAP controls. However, in contrast to the expected results, almost all tested HA-ECM and HA-HAP-ECM fluids exhibited equivalent or increased G′ and τy values compared to respective HA, HAP, and HA-HAP controls (Tables 2 and 3). The contrast between the expected and experimental results can be likely attributed to several of the following phenomena: hydration effects of ECM, intra-particle interactions of ECM, and inter-particle interactions between HAP and ECM. The increase in the aforementioned rheological properties upon the addition of ECM to HA and HA-HAP formulations may be related to the effects of hydration of the incorporated ECM microparticles. As these particles hydrate in the suspending HA solution or HA-HAP colloid, available aqueous media becomes scarcer and carrier fluid concentration increases. This concentrating effect may counteract and outweigh the overall destabilizing contributions caused by stress localization of ECM microparticles. A trend was observed in the HA-ECM microparticle suspensions, wherein G′ increased with increasing ECM concentration (Table 2). ECM concentration dependency was not observed in HA-HAP colloids, suggesting the

ACS Paragon Plus Environment

Langmuir

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 47

24

occurrence of competing fluid phenomenon in these gels. These differences in fluid properties between the HA-ECM suspensions and HA-HAP-ECM colloids may be due to the fact that the total volume fraction (ΦTotal = ΦHAP + ΦECM) of these colloids resided well above (ΦTotal = 0.360.75) the estimated percolation threshold (Φp ≈ 0.22-0.29). At these high concentrations, localized stress experienced by the colloidal fluid may become significant compared to the concentrating effect of the ECM microparticles, leading to complex viscoelastic behavior. In general, it was also observed that formulations containing DBM exhibited higher G′ and τy values than corresponding DCC formulations on both a mass and volume fraction basis (Tables 2 and 3). For some mixtures, an increase in ECM concentration had no statistically significant impact on the rheological properties, specifically for HA-ECM suspensions where a significant increase in both G′ and τy was only observed at the highest ECM volume fraction (ΦECM = 0.45) (p < 0.01). However, HA-HAP-DBM mixtures exhibited significantly higher G′ and τy than all mass fraction equivalent DCC formulations. Some of these observed viscoelastic differences between DBM and DCC formulations could have been attributed to deviations in sample preparation, including fluid component heterogeneity and shear-history. During mixing and reconstitution, samples may have experienced a broad range of shear stresses, and strongly flocculated colloids have been shown to be extremely dependent on shear-history.37,

38

Subsequent recovery studies (Figures 4 and 5), however, concluded that the colloidal ECMbased gels recovered viscoelastic properties very rapidly following microstructure breakdown. Allowing these samples to equilibrate for multiple hours before testing ensured that shear-history effects were minimized. Heterogeneity of suspensions and colloids was also minimized by thoroughly mixing formulations as dry powders prior to reconstitution.

ACS Paragon Plus Environment

Page 25 of 47

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

Langmuir

25

Because the effect of both mass fraction and volume fraction of respective ECM components was controlled in these formulations, differences in water content and density of hydrated DBM versus DCC were accounted for in viscoelastic studies. However, differences in ECM intraparticle properties could have affected overall viscoelastic trends. Pure DBM would likely exhibit significantly higher elastic modulus compared to DCC under compression. Although these measurements were not directly collected in the current study, referenced values for elastic modulus of DBM (130-230 MPa)39 were several orders of magnitude higher than respective DCC values (1.9 MPa).40 Since both HA-ECM and HA-HAP-ECM were formulated near or above Φp, ECM particles formed a continuous disordered network throughout the sample volume, meaning differences in intra-particle elastic modulus could have played a significant role in overall bulk viscoelastic properties. Inter-particle interactions between HAP and ECM particles may also have been responsible for the differences in viscoelastic properties between DBM and DCC formulations. It is possible that HAP interacted differently with DBM upon reconstitution compared to DCC since significant increases in G′ and τy were predominately observed in colloidal gels instead of HAECM formulations. During demineralization, DBM is stripped of biological apatite, which has dimensions on the order of several nanometers.12,

41

The intimate association of apatite

nanoparticles in healthy bone with the organic osteoid matrix, primarily collagen type 1, lead to a significant increase in the overall elastic modulus of the tissue (16-23 GPa).39 It may be possible that HAP nanoparticles present in tested colloidal formulations associated with DBM and subsequently increased G′ and τy. The mechanism of this interaction may be similar to direct mineralization observed in collagen and DBM scaffolds.42 Further work is necessary to elucidate

ACS Paragon Plus Environment

Langmuir

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 47

26

the mechanism of interaction between HAP nanoparticles and ECM microparticles and how it may even be harnessed as a parameter for adjusting the rate degradation for HA-HAP-ECM gels. Finally, the results of the cell study provide a preliminary biological characterization of a wide range of HA-ECM and HA-HAP-ECM formulations (1500 kDa HA). Due to autofluorescence of the gels, it was not possible to quantify and compare the extent of cell death between groups. However, on a qualitative basis, differing amounts of cell death can be noted for each group (Figure 7). For HA-DCC, HA-DBM, and HA-HAP-DBM groups, the number of dead cells appears to decrease with increasing ECM concentration (Supplementary Figures S3 and S4). This effect is markedly visibly over the concentration range between HA-DBM 3% and HA-DBM 28%, with the latter group displaying the fewest dead cells. However, without statistical power, these trends cannot be confirmed. While these data serve as a confirmation that all of the HA-ECM suspensions and HA-HAP-ECM gels that were characterized in the viscoelastic studies are biocompatible formulations that promote cell viability, future work must be conducted in order to identify trends within the biological characterization data. Future work should confirm cell viability over a longer time period, and since autofluorescence posed an issue with the live/dead assay, a different strategy such as a double-stranded DNA quantitation assay should be utilized. In addition to short term biocompatibility, the effects of long-term systemic exposure to both HA-ECM and HA-HAP-ECM fluids must not be overlooked. The toxicity of nanomaterials is a growing concern, since a material that is widely accepted to be inert on the macroscale (e.g., gold), can cause unexpected effects on the nanoscale such as inducing oxidative stress on cells.43 Though biocompatibility studies on commercially available nanoparticle HAP products such as

ACS Paragon Plus Environment

Page 27 of 47

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

Langmuir

27

Ostim® have deemed nano-HAP safe in an injectable therapy,44 future preclinical studies with HA-HAP-ECM gels should still scrutinize potential adverse effects upon long term exposure. In addition to viability, cell differentiation and gene expression will be studied for a range of HA-HAP-ECM formulations to classify the potential therapeutic effects prior to in vivo studies. While DBM has previously demonstrated an ability to promote regeneration of critical-sized bone defects via intramembranous (IM) and endochondral (EC) ossification pathways,45-47 our group has recently demonstrated that DCC as well as devitalized cartilage (DVC) can activate chondrogenic markers that promote cartilage regeneration.48,

49

Furthermore, DCC has

demonstrated promising osteogenic potential, specifically by up-regulation of gene markers indicative of EC ossification50 but so far a majority of research and commercial applications are only targeted towards cartilage regeneration. Future work to define the biological responses induced by HA-HAP-ECM gels could not only correlate DBM or DCC concentrations with the expression levels of osteogenic markers, but also shed more light on the role that DCC plays in promoting the EC ossification pathway in bone healing. Upon further study, if the desired therapeutic effects are not achieved with the currently reported formulations of HA-HAP-ECM gels, the addition of other bioactive molecules to the gels should be the next avenue of research to explore. Osteogenic drugs or chemokines could potentially be loaded in the colloidal gels without affecting the mechanical properties, but vastly enhancing the therapeutic outcome. For instance, the albuminoid protein silk sericin is a protein worth studying in conjunction with the HA-HAP-ECM platform, because it is a mitogenic agent that has been shown in several applications to promote osseointegration by inducing nucleation of HAP.51 In fact, when formulated with natural GAG polymers and HA, the presence of silk sericin caused the formation of interconnected porous networks similar to cartilaginous ECM.51

ACS Paragon Plus Environment

Langmuir

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 28 of 47

28

5. CONCLUSION Overall, the rheological and viscoelastic properties of tested HA-ECM and HA-HAP-ECM mixtures were observed to be primarily dependent on the fluid properties of the respective carrier fluid, either HA or HA-HAP. However, the addition of DCC and DBM resulted in equivalent or higher G′, τy and G′Recovery values compared to control fluids, which directly contradicted our initial hypothesis. While some differences were observed between formulations containing DCC compared to DBM counterparts, the main finding of these studies was that HA-ECM and HAHAP-ECM could be formulated with desirable viscoelastic properties for potential surgical injection into tissue defects. Preliminary biological testing by in vitro cell culture with hUCMSCs confirmed that all of the formulations included in the viscoelastic recovery studies do not induce any cytotoxic responses in the cells. Furthermore, the highest HA-DBM suspension explored in the present study (ΦDBM = 0.45; WDBM = 0.28) possessed a similar microparticle concentration compared to DBX® Paste and Putty commercial products (DBM 26% and 31% w/w) as reported by the supplier MTF/Synthes, which could be used as an external reference point for formulations reported here where significantly higher G′ values were observed in HAHAP-ECM (Table 3). Although higher τy values were observed in the HA-DBM (ΦDBM = 0.45) suspension compared to tested formulations, it is possible for HA-HAP-ECM colloidal gels to be further refined by decreasing the relative amount of suspending media, leading to increases in both G′ and τy. As such, these studies were not designed to cover a comprehensive array of all possible HA, HAP, and ECM fluid combinations. Instead, the current work served to highlight the viscoelastic contribution of HA solutions and HA-HAP colloidal gels containing ECM fillers. Further work with these materials includes the analysis of in vitro gene expression among other biological characterization studies to gain insight on the therapeutic potential and long term

ACS Paragon Plus Environment

Page 29 of 47

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

Langmuir

29

behavior in the presence of cells. Future studies in this space may lead to enhanced rheological and viscoelastic features that could suit a wider variety of bone defect indications including but not limited to use in cranial, oral/maxillofacial, mandibular, pelvic and posterolateral spine reconstruction.

6. ACKNOWLEDGMENTS We acknowledge the National Institute of Dental and Craniofacial Research of the National Institutes of Health (NIH R01 DE022471) and the NIGMS Pre-doctoral Training Grant Program (NIH/NIGMS T32-GM08359). The authors declare no competing financial interest.

ACS Paragon Plus Environment

Langmuir

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 30 of 47

30

7. REFERENCES (1) (2) (3)

(4) (5) (6) (7)

(8)

(9)

(10)

(11)

(12)

(13) (14)

(15) (16) (17)

Guvendiren, M., Lu, H.D., Burdick, J.A., Shear-thinning hydrogels for biomedical applications. Soft matter 2012, 8 (2), 260-272. Gruskin, E.; Doll, B. A.; Futrell, F. W.; Schmitz, J. P.; Hollinger, J. O., Demineralized bone matrix in bone repair: history and use. Advanced drug delivery reviews 2012, 64 (12), 1063-77. Habib, M.; Baroud, G.; Gitzhofer, F.; Bohner, M., Mechanisms underlying the limited injectability of hydraulic calcium phosphate paste. Part II: particle separation study. Acta biomaterialia 2010, 6 (1), 250-6. Verma, N. P.; Sinha, A., Effect of solid to liquid ratio on the physical properties of injectable nanohydroxyapatite. Journal of materials science. Materials in medicine 2013, 24 (1), 53-9. Hing, K. A.; Wilson, L. E.; Buckland, T., Comparative performance of three ceramic bone graft substitutes. Spine J 2007, 7 (4), 475-490. Wang, Q.; Wang, J.; Lu, Q.; Detamore, M. S.; Berkland, C., Injectable PLGA based colloidal gels for zero-order dexamethasone release in cranial defects. Biomaterials 2010, 31 (18), 4980-6. Dennis, S. C.; Detamore, M. S.; Kieweg, S. L.; Berkland, C. J., Mapping GlycosaminoglycanHydroxyapatite Colloidal Gels as Potential Tissue Defect Fillers. Langmuir 2014, 30 (12), 35283537. Wang, H.; Boerman, O. C.; Sariibrahimoglu, K.; Li, Y.; Jansen, J. A.; Leeuwenburgh, S. C., Comparison of micro- vs. nanostructured colloidal gelatin gels for sustained delivery of osteogenic proteins: Bone morphogenetic protein-2 and alkaline phosphatase. Biomaterials 2012, 33 (33), 8695-703. Wang, H.; Zou, Q.; Boerman, O. C.; Nijhuis, A. W.; Jansen, J. A.; Li, Y.; Leeuwenburgh, S. C., Combined delivery of BMP-2 and bFGF from nanostructured colloidal gelatin gels and its effect on bone regeneration in vivo. Journal of controlled release : official journal of the Controlled Release Society 2013, 166 (2), 172-81. Van Tomme, S. R.; van Nostrum, C. F.; de Smedt, S. C.; Hennink, W. E., Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres. Biomaterials 2006, 27 (22), 4141-4148. Wang, Q.; Gu, Z.; Jamal, S.; Detamore, M. S.; Berkland, C., Hybrid hydroxyapatite nanoparticle colloidal gels are injectable fillers for bone tissue engineering. Tissue engineering. Part A 2013, 19 (23-24), 2586-93. Detsch, R.; Hagmeyer, D.; Neumann, M.; Schaefer, S.; Vortkamp, A.; Wuelling, M.; Ziegler, G.; Epple, M., The resorption of nanocrystalline calcium phosphates by osteoclast-like cells. Acta biomaterialia 2010, 6 (8), 3223-33. Song, F.; Li, X.; Wang, Q.; Liao, L.; Zhang, C., Nanocomposite Hydrogels and Their Applications in Drug Delivery and Tissue Engineering. Journal of Biomedical Nanotechnology 2015, 11 (1), 40-52. Benders, K. E.; van Weeren, P. R.; Badylak, S. F.; Saris, D. B.; Dhert, W. J.; Malda, J., Extracellular matrix scaffolds for cartilage and bone regeneration. Trends in biotechnology 2013, 31 (3), 16976. Renth, A. N.; Detamore, M. S., Leveraging "raw materials" as building blocks and bioactive signals in regenerative medicine. Tissue engineering. Part B, Reviews 2012, 18 (5), 341-62. Tsang, K. Y.; Cheung, M. C.; Chan, D.; Cheah, K. S., The developmental roles of the extracellular matrix: beyond structure to regulation. Cell and tissue research 2010, 339 (1), 93-110. Gerstenfeld, L. C.; Cullinane, D. M.; Barnes, G. L.; Graves, D. T.; Einhorn, T. A., Fracture healing as a post-natal developmental process: molecular, spatial, and temporal aspects of its regulation. J Cell Biochem 2003, 88 (5), 873-84.

ACS Paragon Plus Environment

Page 31 of 47

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

Langmuir

31 (18)

(19)

(20) (21)

(22)

(23)

(24) (25)

(26) (27)

(28) (29) (30)

(31) (32)

(33)

(34)

Sfeir, C., Ho, L., Doll, B.A., Azari, K., Hollinger, J.O., Fracture Repair. In Bone Regeneration and Repair: Biology and Clinical Applications, Lieberman, J. R., Friedlaender, G. E., Ed. Humana Press Inc.: Totowa, NJ, 2005; pp 21-44. Mehta, M.; Schmidt-Bleek, K.; Duda, G. N.; Mooney, D. J., Biomaterial delivery of morphogens to mimic the natural healing cascade in bone. Advanced drug delivery reviews 2012, 64 (12), 125776. Discher, D. E.; Mooney, D. J.; Zandstra, P. W., Growth Factors, Matrices, and Forces Combine and Control Stem Cells. Science 2009, 324 (5935), 1673-1677. Converse, G. L.; Armstrong, M.; Quinn, R. W.; Buse, E. E.; Cromwell, M. L.; Moriarty, S. J.; Lofland, G. K.; Hilbert, S. L.; Hopkins, R. A., Effects of cryopreservation, decellularization and novel extracellular matrix conditioning on the quasi-static and time-dependent properties of the pulmonary valve leaflet. Acta biomaterialia 2012, 8 (7), 2722-9. Holland, T. A.; Tabata, Y.; Mikos, A. G., In vitro release of transforming growth factor-beta 1 from gelatin microparticles encapsulated in biodegradable, injectable oligo(poly(ethylene glycol) fumarate) hydrogels. Journal of controlled release : official journal of the Controlled Release Society 2003, 91 (3), 299-313. Ozbas, B.; Rajagopal, K.; Haines-Butterick, L.; Schneider, J. P.; Pochan, D. J., Reversible stiffening transition in beta-hairpin hydrogels induced by ion complexation. The journal of physical chemistry. B 2007, 111 (50), 13901-8. Larson, R. G., The Structure and Rheology of Complex Fluids. Oxford University Press: New York, New York, 1999. Ozbas, B.; Rajagopal, K.; Haines-Butterick, L.; Schneider, J. P.; Pochan, D. J., Reversible stiffening transition in beta-hairpin hydrogels induced by ion complexation. J Phys Chem B 2007, 111 (50), 13901-13908. Shih, W. Y.; Shih, W. H.; Aksay, I. A., Elastic and yield behavior of strongly flocculated colloids. J Am Ceram Soc 1999, 82 (3), 616-624. Mellott, A. J.; Devarajan, K.; Shinogle, H. E.; Moore, D. S.; Talata, Z.; Laurence, J. S.; Forrest, M. L.; Noji, S.; Tanaka, E.; Staecker, H.; Detamore, M. S., Nonviral Reprogramming of Human Wharton's Jelly Cells Reveals Differences Between ATOH1 Homologues. Tissue engineering. Part A 2015, 21 (11-12), 1795-809. Berriaud, N.; Milas, M.; Rinaudo, M., Rheological study on mixtures of different molecular weight hyaluronates. International journal of biological macromolecules 1994, 16 (3), 137-42. Matteini, P.; Dei, L.; Carretti, E.; Volpi, N.; Goti, A.; Pini, R., Structural behavior of highly concentrated hyaluronan. Biomacromolecules 2009, 10 (6), 1516-22. Fakhari, A.; Phan, Q.; Berkland, C., Hyaluronic acid colloidal gels as self-assembling elastic biomaterials. Journal of biomedical materials research. Part B, Applied biomaterials 2014, 102 (3), 612-8. Peng, H.; Liu, X.; Wang, R.; Jia, F.; Dong, L.; Wang, Q., Emerging nanostructured materials for musculoskeletal tissue engineering. Journal of Materials Chemistry B 2014, 2 (38), 6435-6461. Israelachvili, J. N.; Mcguiggan, P. M., Effect of Ions and Surface Lattice Mismatch Angle on Adhesion and Hydration Forces between Mica Surfaces in Electrolyte-Solutions. Abstr Pap Am Chem S 1991, 201, 245-Coll. Leong, Y. K.; Boger, D. V.; Scales, P. J.; Healy, T. W.; Buscall, R., Control of the Rheology of Concentrated Aqueous Colloidal Systems by Steric and Hydrophobic Forces. J Chem Soc Chem Comm 1993, (7), 639-641. Lin, Y.; Boker, A.; He, J.; Sill, K.; Xiang, H.; Abetz, C.; Li, X.; Wang, J.; Emrick, T.; Long, S.; Wang, Q.; Balazs, A.; Russell, T. P., Self-directed self-assembly of nanoparticle/copolymer mixtures. Nature 2005, 434 (7029), 55-9.

ACS Paragon Plus Environment

Langmuir

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 32 of 47

32 (35) (36) (37) (38) (39)

(40)

(41) (42) (43) (44)

(45)

(46)

(47) (48)

(49) (50)

(51)

Garboczi, E. J.; Snyder, K. A.; Douglas, J. F.; Thorpe, M. F., Geometrical Percolation-Threshold of Overlapping Ellipsoids. Phys Rev E 1995, 52 (1), 819-828. Khan, S. A.; Prudhomme, R. K.; Larson, R. G., Comparison of the Rheology of Polymer Melts in Shear, and Biaxial and Uniaxial Extensions. Rheol Acta 1987, 26 (2), 144-151. Buscall, R.; Mcgowan, J. I.; Mortonjones, A. J., The Rheology of Concentrated Dispersions of Weakly Attracting Colloidal Particles with and without Wall Slip. J Rheol 1993, 37 (4), 621-641. Buscall, R.; Mills, P. D. A.; Yates, G. E., Viscoelastic Properties of Strongly Flocculated Polystyrene Latex Dispersions. Colloid Surface 1986, 18 (2-4), 341-358. Hamed, E.; Novitskaya, E.; Li, J.; Chen, P. Y.; Jasiuk, I.; McKittrick, J., Elastic moduli of untreated, demineralized and deproteinized cortical bone: Validation of a theoretical model of bone as an interpenetrating composite material. Acta Biomater 2012, 8 (3), 1080-1092. Schwarz, S.; Koerber, L.; Elsaesser, A. F.; Goldberg-Bockhorn, E.; Seitz, A. M.; Durselen, L.; Ignatius, A.; Walther, P.; Breiter, R.; Rotter, N., Decellularized cartilage matrix as a novel biomatrix for cartilage tissue-engineering applications. Tissue engineering. Part A 2012, 18 (2122), 2195-209. Dorozhkin, S. V., Nanosized and nanocrystalline calcium orthophosphates. Acta biomaterialia 2010, 6 (3), 715-34. Meyer, U., Weismann, H. P. , Bone and Cartilage. In Bone and Cartilage Engineering, Springer: New York, NY, 2006; pp 7-46. Aillon, K. L.; Xie, Y.; El-Gendy, N.; Berkland, C. J.; Forrest, M. L., Effects of nanomaterial physicochemical properties on in vivo toxicity. Adv Drug Deliv Rev 2009, 61 (6), 457-66. Laschke, M. W.; Witt, K.; Pohlemann, T.; Menger, M. D., Injectable nanocrystalline hydroxyapatite paste for bone substitution: in vivo analysis of biocompatibility and vascularization. Journal of biomedical materials research. Part B, Applied biomaterials 2007, 82 (2), 494-505. Wang, J.; Glimcher, M. J.; Mah, J.; Zhou, H. Y.; Salih, E., Expression of bone microsomal casein kinase II, bone sialoprotein, and osteopontin during the repair of calvarial defects. Bone 1998, 22 (6), 621-8. Rabie, A. B. M.; Chay, S. H.; Wong, A. M. K., Healing of autogenous intramembranous bone in the presence and absence of homologous demineralized intramembranous bone. Am J Orthod Dentofac 2000, 117 (3), 288-297. Oakes, D. A.; Lee, C. C.; Lieberman, J. R., An evaluation of human demineralized bone matrices in a rat femoral defect model. Clin Orthop Relat R 2003, (413), 281-290. Beck, E. C.; Barragan, M.; Tadros, M. H.; Kiyotake, E. A.; Acosta, F. M.; Kieweg, S. L.; Detamore, M. S., Chondroinductive Hydrogel Pastes Composed of Naturally Derived Devitalized Cartilage. Annals of Biomedical Engineering 2016, 44 (6), 1863-1880. Sutherland, A. J.; Detamore, M. S., Bioactive Microsphere-Based Scaffolds Containing Decellularized Cartilage. Macromolecular Bioscience 2015, 15 (7), 979-989. Sutherland, A. J.; Beck, E. C.; Dennis, S. C.; Converse, G. L.; Hopkins, R. A.; Berkland, C. J.; Detamore, M. S., Decellularized cartilage may be a chondroinductive material for osteochondral tissue engineering. PloS one 2015, 10 (5), e0121966. Lamboni, L.; Gauthier, M.; Yang, G.; Wang, Q., Silk sericin: A versatile material for tissue engineering and drug delivery. Biotechnology Advances 2015, 33 (8), 1855-1867.

ACS Paragon Plus Environment

Page 33 of 47

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

Langmuir

33

Table of Contents Figure: Schematic depicting the formulation parameters and rheological fluid properties of HA-ECM microparticle suspensions and HA-HAP-ECM colloidal gels. HA, hyaluronic acid; HAP, hydroxyapatite; ECM, extracellular matrix

ACS Paragon Plus Environment

Langmuir

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 34 of 47

34

Table 1: HAP Size and Zeta Potential

*Indicates statistically significant differences from HAP control group (n = 15; p < 0.05). **Indicates statistically significant differences from HAP control group (n = 15; p < 0.01). Data are reported as the mean + standard deviation of the mean (n = 15). HA, hyaluronic acid; HAP, hydroxyapatite.

ACS Paragon Plus Environment

Page 35 of 47

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

Langmuir

35

Table 2: LVE region viscoelastic properties of HA-ECM suspensions

*Indicates statistically significant difference from HA control (p < 0.05). #

Indicates statistically significant difference from HA-DBM Φ = 0.05, (p < 0.05). Data are reported as the

mean + standard deviation of the mean (n = 3). HA, hyaluronic acid; DCC, decellularized cartilage; DBM, demineralized bone matrix.

ACS Paragon Plus Environment

Langmuir

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 36 of 47

36

Table 3: LVE region viscoelastic properties of HA-HAP-ECM gels

*Indicates statistically significant difference from HAP control (p < 0.05). #

Indicates statistically significant difference from HA-HAP Φ = 0.05, (p < 0.05). Data are reported as the mean + standard deviation of the mean (n = 3). HA-HAP-DBM (ΦDBM ≥ 30%) exhibited incomplete hydration and could not be measured. HA, hyaluronic acid; HAP, hydroxyapatite; DCC, decellularized cartilage; DBM, demineralized bone matrix.

ACS Paragon Plus Environment

Page 37 of 47

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

Langmuir

37

Figure 1: SEM images of sieved (< 350 µm) ECM microparticles including A) DBM and B) DCC as well as C) HAP nanoparticles. All three materials displayed polydisperse particle sizes. Notably however, while DBM and DCC had heterogeneous morphologies, the SEM images of the HAP nanoparticles revealed uniformly spherical nanoparticles. Scale bars for images A and B were (100 µm) and for image C (500 nm). ECM, extracellular matrix; DBM, demineralized bone matrix; DCC, decellularized cartilage; HAP, hydroxyapatite.

Figure 2: Swelling ratio (S) of HA-HAP gels plotted versus HAP Φ. Data sets represent HAP colloids containing different molecular weight HA (35, 350, and 1500 kDa) compared to pure HAP. No swelling change was desired (S-ratio ≈ 0), and success criteria for the gels were established by setting S-ratio tolerances (dashed lines; 0 ± 20%). Individual points are reported (average ± SD) from triplicate studies. HA, hyaluronic acid; HAP, hydroxyapatite.

Figure 3: LVE region viscoelastic properties from controlled stress sweeps (1-10,000 Pa) at constant frequency (1 Hz) of HA-HAP mixtures composed of 35 kDa (10% w/v), 350 kDa (3% w/v), and 1500 kDa (1% w/v) HA polymers and compared to HAP and HA control fluids, respectively. Averaged values for G′, G″, and τy (Pa) are displayed from triplicate studies. G’ and G’’ are displayed on the primary axis, while τy is displayed on the secondary axis. Error bars represent the standard deviation from the average of triplicate studies; statistically significant differences from HA (35 kDa) control group (*p