Effect of Assembly pH on Polyelectrolyte Multilayer ... - ACS Publications

May 17, 2016 - and Amy M. Peterson*,†,‡,§. Departments of. †. Chemical Engineering,. ‡. Mechanical Engineering, and. §. Biomedical Engineeri...
0 downloads 0 Views 2MB Size
Subscriber access provided by University of Sussex Library

Article

Effect of assembly pH on polyelectrolyte multilayer surface properties and BMP-2 release Claire Salvi, Xuejian Lyu, and Amy M Peterson Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.5b01730 • Publication Date (Web): 17 May 2016 Downloaded from http://pubs.acs.org on May 18, 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.

Biomacromolecules 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 40

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

Biomacromolecules

Effect of assembly pH on polyelectrolyte multilayer surface properties and BMP-2 release Claire Salvi1, Xuejian Lv2, Amy M. Peterson1,2,3* 1

Department of Chemical Engineering, Worcester Polytechnic Institute, 100 Institute Road,

Worcester, MA 01609 2

Department of Mechanical Engineering, Worcester Polytechnic Institute, 100 Institute Road,

Worcester, MA 01609 3

Department of Biomedical Engineering, Worcester Polytechnic Institute, 100 Institute Road,

Worcester, MA 01609

ABSTRACT The effect of solution pH during layer-by-layer buildup of polyelectrolyte multilayer (PEM) coatings on properties relevant to orthopedic implant success was investigated. Bone morphogenetic protein 2 (BMP-2), a potent osteoconductive growth factor, was adsorbed onto the surface of anodized titanium, and PEM coatings prepared from solutions of poly-Lhistidine and poly(methacrylic acid) were built on top of the BMP-2. High levels of BMP-2 released over several months were achieved. Approximately 2 µg/cm² of BMP-2 were initially adsorbed on the anodized titanium and a pH-dependent release behavior was observed, with more stable coatings assembled at pH = 6-7. Three different diffusion regimes could be determined from the release profiles: an initial burst release, a sustained release regime and a

ACS Paragon Plus Environment

1

Biomacromolecules

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 40

depletion regime. BMP-2 was shown to maintain bioactivity after release from a PEM and the presence of a PEM was shown to preserve BMP-2 structure. No visible change was observed in surface roughness as the assembly pH was varied, whereas the surface energy decreased for samples prepared at more basic pH. These results indicate that the initial BMP-2 layer affects PEM surface structure, but not the functional groups exposed on the surface.

INTRODUCTION Orthopedic implanted devices are used to relieve or replace deficient articulations such as knees, hips, and the spine. In 2012, over 1.55 million knee replacements, hip replacements and spinal fusions were performed in the United States alone, which represented almost 20% of all surgeries registered that year.1 All three of these procedures have been consistently ranked in the top ten most performed as well as the top ten most expensive procedures.2,3 One of the major concerns regarding implants is to ensure their long term performance. Nonetheless, more than 100,000 revisions to knee and hips implants were performed in 2012, which roughly represents a 10% revision rate per year. This revision rate has been overall stable since 1994.4 One of the leading causes of implant revision is limited integration with the surrounding bone tissue, either because of poor adhesion/growth of bone cells and tissue on the scaffold or because of micro-motions inhibiting the biological integration process. Research has thus been focused on ways of increasing the early osseointegration of implants in order to increase their lifespan. Strategies including radiation therapy, bone marrow grafting, demineralized bone matrix use, and local delivery of biologics such as growth or differentiation factors to the injury site have been proposed.5 In the area of local growth factor delivery, vascular endothelial growth factor

ACS Paragon Plus Environment

2

Page 3 of 40

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

Biomacromolecules

(VEGF),6 fibroblast growth factors (FGFs),7 and bone morphogenetic proteins (BMPs)8,9 have been investigated for orthopedic applications. Although devices containing BMP-2 and BMP-7 have been approved by the American Food and Drug Administration (FDA) to increase osteoblast differentiation as well as bone extracellular matrix mineralization, current delivery methods and doses have been shown to cause adverse side effects such as osteolysis, ectopic bone growth, and a greater risk of developing cancer.10 Research has thus focused on lower levels of sustained release of BMP-2 in order to prevent these side effects. One option is a growth factor-eluting coating applied to the surface of the implant. Various designs have been proposed such as fibroin microparticles,11 hydrogels,12 and polyelectrolyte multilayers.13–18 Polyelectrolytes are polymers with ionically dissociable repeat units. These groups are positively charged for polycations, negatively charged for polyanions, or including both positive and negative charges for polyampholytes. The electrostatic interactions between the different charged groups can lead to the formation of polyelectrolyte complexes, for example by alternating layer-by-layer adsorption of a polycation and a polyanion. This specific structure, called a polyelectrolyte multilayer (PEM), can be designed as a thin film or as capsules. Since many polyelectrolytes have shown excellent biocompatibility, PEMs have been used in numerous biomedical applications. Examples include controlling cell transfection or differentiation,19,20 obtaining nano-patterned surfaces,16,21 design of a superhydrophobic surface,22 functionalization of living cells,23 and delivery systems for biologically relevant molecules including DNA,24 antibiotics and anti-inflammatory drugs,25–27 and growth factors.7,14,15,17,28,29

ACS Paragon Plus Environment

3

Biomacromolecules

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 40

PEMs have previously been used for the controlled release of BMP-2 from orthopedic implant surfaces. Macdonald et al. demonstrated the coating of a polymer scaffold with LbL-deposited poly(β-aminoester) (PBAE) and chondroitin sulfate (CS), a complex capable of delivering microgram scale amounts of BMP-2.14 Less than 1% of release occurs in the first three hours, with 80% of BMP-2 released in the first two days and full release achieved in two weeks. Enhanced preosteoblast differentiation was observed in the BMP-2 releasing system as compared to similar concentrations of BMP-2 added directing to the culture medium, suggesting a synergistic effect of BMP-2 and the PEM. Another form of this system, capable of combined BMP-2 and VEGF-C release, resulted in 33% higher bone density in de novo formed bone, as compared to scaffolds with just BMP-2 release.30 Crouzier et al. prepared cross-linked poly-Llysine (PLL)/hyaluronic acid (HA) PEMs on a porous ceramic scaffold that were also capable of microgram level release of BMP-2.13 Burst release over the first three days represented 30% of the loaded BMP-2 for the high concentration cross-linker condition and 75% for the low concentration cross-linker condition. While BMP-2 bioactivity was demonstrated in vitro, in vivo osteoinduction was not enhanced on the PEM-coated scaffolds. Subsequent studies of the crosslinked PLL/HA coating on titanium surfaces showed that the secondary structure of BMP-2 is maintained in hydrated and dry coatings.28,31 PEM properties and functionality are dependent on many processing and environmental variables. Over the past decade, several reviews have focused on how stimuli such as external pH, temperature, electric or magnetic fields, salt or sugar concentration can impact PEM permeability, thickness, swelling behavior, drug loading capacity, surface roughness or cell adhesion.32–36 Further control over the release from these systems can be achieved by either cross-linking the different layers or including nanoparticles in its structure.8,13,37,38 Thompson et

ACS Paragon Plus Environment

4

Page 5 of 40

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

Biomacromolecules

al. showed that the assembly pH was as important as the external pH in determining PEM properties such as elasticity and cell adhesion.39 Peterson et al. reported similar observations for another BMP-2-eluting PEM.40 This particular system consisted of a BMP-2 layer adsorbed on an anodized titanium substrate and covered with five bilayers of poly(methacrylic acid) (PMAA) and poly-L-histidine (PLH), denoted as (PMAA/PLH)5. Sustained release of BMP-2 had been previously observed for several weeks;17 however, adjusting the assembly pH to 6.0 greatly increased the eluted amount of BMP-2. In the current work, PEM assembly pH was systematically studied in order to investigate its impact on the properties of the resulting coating. The novel aspect of this work is the investigation of the effect of PEM assembly pH on BMP-2 release and surface properties. The PEM considered was made of an initial layer of BMP-2 adsorbed on anodized titanium and covered with (PMAA/PLH)5. Specifically, the effects of the PEM assembly pH on BMP-2 release from the PEM and surface properties of the PEM were investigated. Potential BMP-2 release mechanisms were also explored based on the analysis of release kinetics.

EXPERIMENTAL SECTION Materials. Titanium foil (99.5% metal basis, 0.25 mm thick) and titanium wire (99.7% metal basis, 0.25 mm in diameter) were purchased from Alfa Aesar. Poly(methacrylic acid, sodium salt) solution (PMAA, Mn ~ 5400), poly-L-histidine hydrochloride (PLH, molecular weight ≥ 5000), phosphate buffered saline (PBS, pH = 7.4), sodium hydroxide anhydrous pellets (≥ 98%), hydrochloric acid and sulfuric acid (95-98%) were obtained from Sigma-Aldrich. Recombinant human bone morphogenetic protein 2 (BMP-2) as well as human BMP-2 enzyme linked immune sorbent assay (ELISA) development kits were acquired from Peprotech.

ACS Paragon Plus Environment

5

Biomacromolecules

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 40

Titanium preparation. Titanium foil plates were cleaned in 1.5 M sulfuric acid and then rinsed in water, ethanol, acetone, and water again. The plates were then anodized in a 165 g/L sulfuric acid solution at a potential of 30 V for 5 minutes and rinsed afterwards with deionized water. Under those conditions, the pore sizes of the titanium dioxide surface range from 40 to 200 nm in diameter.41 PEM coating preparation. Each anodized titanium plate was immersed in a 100 µg/mL BMP-2 in water solution for 15 minutes, during which time the BMP-2 adsorbed to the anodized titanium surface. After the deposition of this preliminary layer, the plates were rinsed three times in deionized water and the PEM was built on top of it as follows. First, the plates were immersed in a 1 mg/mL PMAA in water solution for 15 minutes, after which they were rinsed three times in deionized water in order to remove loosely adsorbed polyelectrolyte from the surface. They were then immersed in a 1 mg/mL PLH in water solution and once again left to adsorb for fifteen minutes before being rinsed three times in deionized water, thus concluding the formation of the first polyelectrolyte bilayer. This process was repeated until five PMAA/PLH bilayers (ten layers total) were obtained. Five bilayers were selected because Peterson et al. showed that five PMAA/PLH bilayers were sufficient for sustained release of BMP-2.17 Samples were dried with a stream of nitrogen and stored in a refrigerator at 4 °C overnight. For a given PEM, the PMAA and PLH solution as well as the water used for rinsing were adjusted to the same pH using hydrochloric acid and/or sodium hydroxide solutions. Coated anodized titanium plates were prepared using solutions adjusted to pH 4.0, 5.0, 6.0, 7.0 and 8.0 in order to study the impact of assembly pH on BMP-2 release and the coating structure. Monitoring of the pH adjustment showed that the concentration of sodium/chloride ions added to

ACS Paragon Plus Environment

6

Page 7 of 40

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

Biomacromolecules

the solutions was relatively constant and far less than 10-6 M; therefore, the influence of ions on polyelectrolyte conformation and resulting PEM structure was deemed negligible. Growth factor release. For each pH condition, the BMP-2 release study was conducted on five PEM-coated anodized titanium samples. After immersion in PBS, the plates were incubated at 37 °C. Aliquots of 1 mL were regularly sampled, with the aliquot volume replaced with fresh PBS. All aliquots were immediately frozen at -30 °C. The amount of eluted BMP-2 was quantified using a sandwich enzyme linked immunosorbent assay (ELISA). ELISA was performed in accordance with the instructions provided with the development kit. Aliquots from the release studies were thawed and returned to room temperature immediately prior to their use. Color monitoring of the ELISA plates was performed using a Perkin-Elmer Victor3 multilabel reader with a 405 nm filter and 650 nm correction filter. Quartz crystal microbalance. Quartz crystal microbalance with dissipation monitoring (QCM-D) was used to measure changes in PEM mass over time. QCM-D was performed using a Q-Sense E4 on titanium sensors at a flow rate of 50 µL/min. PEMs were assembled at 22 °C and degradation studies were performed in PBS at 37 °C. Water viscosity and density are temperature dependent, so changes in frequency and dissipation as measured by QCM-D are also temperature-dependent.42 To separate frequency shifts due to changes in temperature from frequency shifts due to changes in the mass adsorbed to a QCM-D sensor, empty sensors underwent the same temperature changes as PEM-coated sensors.43,44 The empty sensor response was subtracted from the PEM-coated sensor response to provide corrected frequency data. The frequency shift was correlated to changes in mass through the Sauerbrey equation: 

∆ = ∆

(1)

ACS Paragon Plus Environment

7

Biomacromolecules

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 40

In this equation, m is the adsorbed mass, f the resonant frequency, n the overtone number and C the mass sensitivity constant, which in this case is 17.7 ng/(cm2 Hz). BMP-2 bioactivity assay. BMP-2 has been shown to induce osteoblastic differentiation in C2C12 myoblasts when in concentrations of 300 ng/mL or higher.45 The bioactivity of BMP-2 after release from PEMs was assessed based its ability to induce osteoblastic differentiation in C2C12 myoblasts.31,46 C2C12 mouse myoblasts obtained from the American Type Culture Collection (ATCC) were cultured in a 1:1 (v/v) ratio of high glucose Dulbecco’s modified Eagle medium (DMEM, Gibco) supplemented with 4 mM L-glutamine and Ham’s F12 (Gibco) with 10% fetal bovine serum (FBS, Sigma-Aldrich), 1% amphotericin B (250 µg/mL, Thermo Fisher) and 1% penicillin streptomycin (10,000 U/mL, Thermo Fisher). Cells were incubated at 37 °C with 5% CO2 and maintained at a density below 70% confluence using standard cell culture techniques. Routine cell passage was conducted using 0.25% trypsin–EDTA (Mediatech).47 The bioactivity of BMP-2 was determined by measuring the alkaline phosphatase (ALP) enzyme activity, which is a marker of osteoblastic differentiation, of C2C12 myoblasts after culture on BMP-2 eluting surfaces.45,46,48 Anodized titanium specimens (1 x 1 cm2) were placed individually in the wells of 24-well plates and C2C12 myoblasts were seeded at a density of 90,000 per well in 1 ml of growth medium. After 3 days, the medium was removed and the specimens were washed with PBS. Cells were then lysed using sonication in 500 µm of 0.1% Triton-X100. ALP enzyme activity was measured using an alkaline phosphatase colorimetric assay kit (Abcam). Atomic force microscopy. The surfaces of PEM-coated titanium plates prepared as described previously were characterized using an atomic force microscope (AFM, Nanosurf NaioAFM) in contact mode. For each AFM image, the root mean square (RMS) roughness of the surface as

ACS Paragon Plus Environment

8

Page 9 of 40

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

Biomacromolecules

well as the peak-valley height were measured and then averaged by using three samples per pH condition. Contact angle analysis. Contact angle analysis was performed using a contact angle goniometer (Ramé-Hart). For each condition, 2 µL droplets of four known liquids (water, glycerol, formamide, and diiodomethane) were placed onto the corresponding plate and the resulting contact angles were measured and recorded. At least six droplets per liquid per surface were measured. The surface energy of each PEM coating was determined using the OwensWendt theory.49 Statistical analysis. Error bars indicate standard deviations. When applicable, one way analysis of variance (ANOVA) was used to determine statistical significance. Tukey’s HSD test was also performed to determine which conditions’ means were statistically different.

RESULTS AND DISCUSSION BMP-2 Release – Effect of Deposition Conditions. BMP-2 release results from PEMs assembled at different pH values are shown in Figure 1. For PEMs prepared at the lowest pH (pH = 4.0), the amount of BMP-2 released over 25 days is the greatest (1.80 ± 0.11 µg/cm²). As the assembly pH becomes more basic this amount decreases gradually, reaching its minimum release over 25 days for coatings assembled between pH = 6.0 and 7.0 with roughly 790 ± 160 ng/cm² of BMP-2 released in the same time span. However, when the deposition pH was increased further, the amount of BMP-2 released from the PEM increased, with 1.20 ± 0.085 µg/cm² of cumulative release over 25 days for plates prepared at pH = 8.0.

ACS Paragon Plus Environment

9

Biomacromolecules

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 40

Figure 1. Cumulative released BMP-2 from PEMs with various assembly pH values. Error bars represent standard deviation, n = 5. The range of release profiles results from changes in the PEM internal structure as a result of assembly pH. Since the internal pH of all samples quickly reached the pH of the surrounding medium (pH = 7.4), the different PEM structures and resulting release behaviors result from the amount of polyelectrolyte adsorbed during assembly and the polyelectrolytes’ assembly conformation. The deposition pH-dependent variations in internal organization were reflected in the BMP-2 release profiles. When weak polycations and/or polyanions are deposited on a surface in an alternating fashion, electrostatic interactions form a pH-sensitive PEM that exhibits a “closed” state when the opposing charges of the polyelectrolytes neutralize each other, but an “open” state when the environment provides additional charges to neutralize either the polyanion or polycation. Polyanions and polycations in PEM films and microcapsules are often highly entangled with each other, which allows for closing and opening of films several times.50–52 Based on this understanding of BMP-2 release, a smaller amount of BMP-2 released after a certain time span would indicate that the PEM coating is more stable and/or packed more tightly than for other pH conditions. Peterson et al. have shown that PMAA/PLH is insoluble in water

ACS Paragon Plus Environment

10

Page 11 of 40

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

Biomacromolecules

over the range pH = 4.9-9.0 and PMAA/PLH with a higher molecular weight PMAA is insoluble in water over the range pH = 5.7-7.7.41 These results might explain the increased resistance to BMP-2 elution of the studied PEM prepared at pH = 6.0 and 7.0. When a polyelectrolyte complex is insoluble, there is sufficient electrostatic bonding between the polyelectrolytes to cause agglomeration and precipitation of the complex. Therefore, when polyelectrolyte complexes exhibit insolubility, the PEMs of the same polyelectrolytes will exhibit more electrostatic bonding, which will reduce the diffusivity of BMP-2. Extended release studies out to 70 days were performed for the PEMs assembled from pH = 4.0 and pH = 5.0 solutions. Results are shown in Figure 2a. Very little BMP-2 was measured in aliquots for each time point over the final weeks, indicating that the PEM was depleted of BMP2 after 70 days. This assumption was furthermore supported by the nearly identical values of cumulative BMP-2 release attained at the end of both release studies, establishing that the total amount of BMP-2 adsorbed on each plate was 2.0 ± 0.13 µg/cm2. Since BMP-2 adsorption was identical for all PEMs, the amount of BMP-2 contained within each system should be the same. These extended release studies were motivated in part by inconsistent results from studies measuring the amount of BMP-2 left on substrates after 25 days (Supporting Information, Table S1 and Figure S1). Independent of the assembly pH, all samples presented sustained release for up to 25 days. Based on the total amount of BMP-2 adsorbed given above (2.0 µg/cm2), the percentage of total release after 25 days is 90%, 84%, 40%, 36%, and 60% for assembly pH values of 4.0, 5.0, 6.0, 7.0, and 8.0, respectively. The osseointegration process begins almost immediately, with evidence of new bone formation by day 5. While full repair can take up to three years, integration of an implant device with mature bone through newly developed bone is observed

ACS Paragon Plus Environment

11

Biomacromolecules

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 40

after 12 weeks.53,54 BMP-2 eluting coatings have the potential to speed up the osseointegration process.

Figure 2. Cumulative BMP-2 released from PEMs assembled at pH = 4.0 and pH = 5.0 in terms of a. mass and b. percentage. Error bars represent standard deviation, n = 5. The total amount of BMP-2 released at the end of the experiment was identified to be the total amount of BMP-2 initially adsorbed. Previous studies of BMP-2 release demonstrated that high amounts can be released depending on the PEM formation conditions and on the release conditions. Guillot et al. as well as Crouzier et al. used a PEM consisting of PLL and HA and observed concentrations ranging from 1-15

ACS Paragon Plus Environment

12

Page 13 of 40

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

Biomacromolecules

µg/cm² of growth factor being released from the PEM. Crouzier et al. used TCP/HAP granules as substrates, whereas Guillot et al. used titanium.13,28 MacDonald et al. reported as much as 10 µg/mm3 of BMP-2 released in vivo from a 14 mg 3D-printed polymer scaffold coated with a poly(β-aminoester)/chondroitin sulfate multilayer.14 While the substrates and BMP-2 loading procedures for these previous studies are different than those in the current study, they indicate that the amounts of release reported in this study are plausible. The results of this report tend to confirm that several complex phenomena oversee the release process. These results also demonstrate the potential of this PEM coating as a long-term delivery system, as its effectiveness could be witnessed for over two months. As a comparison, BMP-2 release from uncoated surfaces was also measured. BMP-2 was adsorbed to anodized titanium surfaces as described previously. However, no PEM was assembled on the BMP-2. Results are shown in Figure 3. The amount of measured BMP-2 release over four days from an uncoated surface is two orders of magnitude less than BMP-2 release over the same amount of time from PEMs. Since the same protocol is used to apply BMP-2, there should be the same amount of BMP-2 in the PEM and uncoated cases. However, significantly less BMP-2 is measured in the uncoated case. A likely explanation for the disparity lies in the way that BMP-2 release is quantified. ELISA antibodies are designed to measure bioactive BMP-2. While this does not mean that ELISA only measures bioactive BMP-2, it does mean that it may not measure BMP-2 that is physically or chemically degraded. Therefore, the vast difference in measured BMP-2 between the coated and uncoated case indicates that BMP2’s structure is better preserved by PEMs. This result is in agreement with Gilde et al. and Guillot et al.28,31

ACS Paragon Plus Environment

13

Biomacromolecules

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 40

Figure 3. Cumulative BMP-2 released from an uncoated anodized titanium surface. Error bars represent standard deviation, n = 4. BMP-2 Release – Kinetics. From Figure 1 it can be seen that BMP-2 release is highest over the first few days. Constant release was observed after the initial burst release until the end of the 25 day period. The data in Figure 2 complete this observation by covering a longer time span. From Figure 2, the second (linear) release regime lasts in total for 20-30 days. After that point the release rate approaches zero in the third release regime. The first zone corresponds to burst release, a common phenomenon in drug release profiles in polymeric systems. Although it has been observed for numerous PEMs, the underlying mechanisms are still poorly understood. Parameters that may influence diffusion include the substrate surface properties, the morphology and porosity of the PEM coating as well as the processing conditions.55 In the experiments reported here, the immersion of the samples into the release medium could be the main cause of the observed burst release, since the dry PEM coatings swell as water diffuses between the polymer chains.32 This swelling causes chain reorganization, breaking electrostatic bonds and forming new ones as a PEM structure is achieved that is more loosely bound than before due to its hydration. This less dense PEM along

ACS Paragon Plus Environment

14

Page 15 of 40

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

Biomacromolecules

with faster transport due to the presence of water molecules in the multilayer would greatly promote the diffusion of BMP-2. Moreover, a concentration gradient would be established between the BMP-2-covered titanium substrate and the BMP-2-free medium in which the sample is immersed, resulting in osmosis, which would further increase the release rate. The second regime presents a fairly linear profile (R2 ≥ 0.91) over the time range of one to several weeks. The transition to this regime occurs when steady-state diffusion is reached after completion of the PEM swelling and reorganization. This stage of the release is the most interesting one from a medical point of view, as it would allow for a controlled and continuous delivery of growth factor over a long period of time. Moreover, the differences in speed of release (Table 1) and in transition time (Table 2) induced by the assembly pH of the coating indicate that these parameters could be optimized based on assembly pH, thus resulting in precisely tuned BMP-2-eluting PEMs that would be specifically designed to provide a given release profile. Linear release profiles for (PMAA/PLH)5 PEMs were previously reported by Peterson et al.40; however, the considerably longer period of steady release as well as the larger amount of growth factor involved that are reported here further increase the potential of this system. The third regime, in which the rate of release decreases substantially, likely corresponds to the depletion of BMP-2. As the amount of BMP-2 remaining on the plate decreases, constant diffusive flux cannot be maintained through the PEM. The flux of BMP-2 decreases as the adsorbed BMP-2 becomes depleted. The transitions between each regime are reported in Table 2. Not all PEMs reached this third regime in the release time studied.

ACS Paragon Plus Environment

15

Biomacromolecules

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 40

Table 1. Slope of the different kinetic regimes of release. Slope (ng/cm2*day) pH

Regime II

Regime III

4.0

75.8

7.74

5.0

37.9

3.69

6.0

26.9

-

7.0

9.28

-

8.0

19.6

-

Table 2. Transition points between the different kinetic regimes of release. Transition I/II

Transition II/III

pH

t (d)

BMP-2 released t (d) (ng/cm²)

BMP-2 released (ng/cm²)

4.0

0.353

904

10.1

1650

5.0

3.21

899

27.6

1820

6.0

2.67

218

-

-

7.0

3.78

514

-

-

8.0

3.98

803

-

-

Degradative phenomena were deemed to be negligible as compared to diffusive phenomena in the above discussion. To evaluate that assumption, mass loss from a (PLH/PMAA)5.5 assembled at pH = 4.0, 6.0 and 8.0 was measured continuously with QCM-D. Results are summarized in Figure 4. Since (PLH/PMAA)5.5 PEMs have a layer of PLH in place of BMP-2, any loss in mass results from physical degradation of the PEM. For PEMs assembled at pH = 6.0, a substantial drop in mass is observed over the first hour in PBS, followed by a leveling off and gradual

ACS Paragon Plus Environment

16

Page 17 of 40

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

Biomacromolecules

increase in mass. However, PEMs assembled at pH = 4.0 exhibit a sharp drop in mass that is quickly recovered, followed by a gradual increase in mass over the rest of the study. Since the difference in behavior was observed over short time scales, this experiment was run again with (PLH/PMAA)5.5 assembled at pH = 8.0. Results comparing the changes in mass over 12 hours are shown in Figure 4b. The behavior of a PEM assembled at pH = 8.0 has the same shape, but different amplitude as that assembled at pH = 6.0. This difference in maximum mass loss could be due to the PEM at pH = 8.0 being much less massive than the PEM assembled at pH = 6.0; indeed, the mass loss at pH = 8.0 represents almost complete mass loss, as compared to a 28% decrease in the mass of the PEM for pH = 6.0. The short time scale drop in mass could be an artifact from changing the liquid pH or ionic strength, or it could be a real measurement of mass loss during PEM rearrangement as discussed above. The gradual increase observed over longer time scales for all PEMs is most likely the result of bound water contributing to the measured mass, which confirms the PEM swelling described above during the first release regime. Based on these results, degradative phenomena may be significant, depending on assembly pH. These differences in amount of physical degradation may contribute to the lack of trend observed in the transition between kinetic regimes I and II. If degradative phenomena are significant, there could also be implications on the rate and amount of time of steady-state diffusion. Indeed, if the majority of the coating is removed over a short time period, diffusion length will be decrease, causing an increase in rate and a shorter amount of time before the PEM is depleted of BMP-2. Such a scenario may explain the increased BMP-2 release rate from PEMs assembled at pH = 8.0. It should also be noted that assembly pH impacts the amount of polyelectrolyte adsorbed, so PEMs with the same number of bilayers assembled at different pH

ACS Paragon Plus Environment

17

Biomacromolecules

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 40

values do not necessarily have the same mass or thickness. An example of this is shown by the initial PEM mass values in Figure 4.

Figure 4. (PH/PMAA)5.5 mass under release study conditions. a. PEM mass; b. Change in PEM mass. (PH/PMAA)5.5 PEMs were assembled at pH = 4.0, 6.0 or 8.0 on titanium sensors. Release study conditions were 37 °C in a PBS solution. Error bars represent standard deviation, n = 4. Despite the presence of three different regimes for all assembly pH conditions, the variations in regime transition and speed of growth factor release still could not be predicted. More general models taking into account the whole release process were also investigated in order to better understand and control the BMP-2 release profiles. The Higuchi model developed in 1961 is a widely used equation that describes the release rate of drugs from a solid matrix. It can be written as:  

= √

(1)

where Mt is the total amount of drug release at time t, M∞ is the total amount of drug release, and K is a constant reflecting the design variables of the system.56 The Higuchi model is derived from Fick’s first law and both describe transport as a function of the square root of time, so they are only valid for the first 60% of the total drug release.57 Nonetheless, previous studies have had

ACS Paragon Plus Environment

18

Page 19 of 40

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

Biomacromolecules

success describing release from PEMs and hydrogels using such models under specific conditions.40,58 The data obtained during release studies were plotted and fitted to the Higuchi model for all points up to 60% of the total amount released at the end of each experiment, and the resulting model parameters were computed in Table S2. It could be concluded from these results that, in general, this model does not properly predict BMP-2 release from the prepared PEMs. This outcome is unsurprising for the following reasons: 1) The Higuchi model assumes that the polymer carrier diffusing into the outside medium should not swell or degrade during the diffusion process;57 2) Fickian diffusion is better suited to describe the diffusion of small, uncharged molecules instead of charged macromolecules with complex secondary structure.59 Other models were also considered such as a zero-order kinetic behavior and the power law model, which can be written as:  = .  

(2)

Higuchi and Fickian models are obtained when n = 0.5, whereas the zero-order model corresponds to the case where n = 1, which means that the drug release would remain constant over time. Berg et al. found that the zero-order model was consistent with complete drug release profiles for PEM systems;60 however, in most cases a zero-order model was only representative of a later stage in the release process.25,58 From Figure 1, it can be concluded that this model cannot be used to account for the complete release profile. BMP-2 release from each PEM was fit to a power law model. The results are provided in Table S2. Overall, all pH conditions agreed fairly well with a power law, with coefficients of determination greater than 0.87 and powers below 0.5. Extending the model to the release studies lasting seventy days confirmed these results for pH = 4 and pH = 5, with coefficients of determination equal to 0.92 and 0.96 and power orders equal to 0.246 and 0.397, respectively.

ACS Paragon Plus Environment

19

Biomacromolecules

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 40

From this analysis, it can be concluded that, if a Fickian or Higuchi diffusion mechanism is indeed involved in this release process, it is not the only one actively participating in the transport of BMP-2 from the substrate to the outside medium. Siepmann and Peppas reported that the increased plasticity of the multilayer caused by hydration of the polymer matrix can enhance molecule transport through a case-II mechanism.57 The phenomena of diffusion through the PEM, transport induced by dynamic swelling, and physical degradation of the PEM should be all taken into account in order to devise an accurate model of drug release from PEMs. Additionally, the effect of assembly pH on PEM swelling can also change each phenomenon’s relative contribution to transport. Finally, the electrostatic interactions between BMP-2 molecules and the polymer matrix cannot be neglected, as such interplay is highly important during the release process.61 BMP-2 Bioactivity. The bioactivity of growth factors within PEMs is an important consideration for application of such coatings. Although ELISA antibodies are designed to recognize bioactive standards, this does not necessarily mean that the antibodies will not recognize non-bioactive BMP-2 as well. Bioactivity was evaluated by the ability of BMP-2 released from PEMs to induce osteogenic differentiation of myoblasts as measured by ALP enzyme activity. Results for release from BMP-2 containing PEMs assembled at pH = 4.0 are shown in Figure 5 and are compared to three controls: the same PEM without BMP-2, anodized titanium, and anodized titanium with 2 µg BMP-2 added to the growth medium. This amount of BMP-2 was the total amount of BMP-2 released for a PEM assembled at pH = 4.0. This relatively acidic condition was selected because it exhibited the highest BMP-2 release after 3 days, and the osteogenic differentiation of myoblasts is dose dependent. Katagiri et al. reported marked increases in ALP activity from myoblasts in the presence of 300 ng/mL BMP-2 and a

ACS Paragon Plus Environment

20

Page 21 of 40

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

Biomacromolecules

fivefold increase in ALP activity in the presence of 1000 ng/mL BMP-2 after six days of culture.45 Since the culture time used in this work was three days, it was decided to target higher BMP-2 concentrations.

Figure 5. Alkaline phosphatase enzyme activity of C2C12 cells cultured on different surfaces. Ti is an anodized titanium surface; Ti + BMP-2 is an anodized titanium surface, 2 µg BMP-2 was added to the growth medium at time t = 0; pH = 4 is a (PH/PMAA)5.5 PEM assembled on anodized titanium at pH = 4.0; pH = 4 + BMP-2 is BMP-2-(PMAA/PH)5 assembled on anodized titanium at pH = 4.0. Error bars represent standard deviation, n = 3. Only the PEM releasing BMP-2 and the titanium with added BMP-2 demonstrated ALP enzyme activity. Ti + BMP-2 condition exhibited greater ALP enzyme activity at a statistically significant level (p < 0.001). This large difference in activity may be due to a combination of the following situations: 1. More BMP-2 exposure in the Ti + BMP-2 condition (2 µg BMP-2 is added to the growth medium at the start of culture for Ti + BMP-2 vs. 2 µg BMP-2 is immobilized in the PEM; 2. A potential loss of BMP-2 bioactivity in the PEM; 3.Differences in exposure to BMP-2 in the two systems (all at t = 0 for Ti + BMP-2 vs. released continuously for the PEM). These results indicate that the BMP-2 released from these PEMs maintains at least

ACS Paragon Plus Environment

21

Biomacromolecules

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 40

some of its bioactivity, even when the BMP-2 is exposed to relatively acidic conditions (pH = 4.0) during BMP-2 assembly. It may seem curious that Figure 5 displays significant BMP-2 bioactivity in the presence of anodized titanium, while Figure 3 shows almost no detectable BMP-2 release from an anodized titanium surface. However, the samples characterized in Figure 3 were prepared in the same manner as PEM samples, which includes drying under a nitrogen stream followed by storage overnight at 4 °C. Under these conditions, unprotected BMP-2 begins to degrade. The conditions for the bioactivity assay differed: BMP-2 was added directly to the growth medium at the start of the cell culture. Although Figure 5 shows that BMP-2 may lose some bioactivity when incorporated within PEMs, Figure 3 shows that using unprotected (uncoated) BMP-2 on a titanium surface is not practical. AFM. The surface roughness of PEMs designed as possible implant coatings is highly important because surface topography is a key parameter for cell function. Proliferation of some cells, including pre-osteoblasts and fibroblasts, was enhanced by smoother PEM surfaces, whereas differentiation of those cells was promoted by rougher PEM surfaces.40,62,63 Parameters influencing PEM roughness include the number of layers, the nature of the polyelectrolyte deposited as a last layer, the assembly temperature, and even the assembly pH.64–66 Contact mode AFM of the PEM surfaces revealed that the differences between all samples are negligible. Independently of their assembly pH, they all present a fairly flat topography with a few irregularities probably due to locally coiled polymer chains or substrate defects (Supporting Information). This observation is reinforced by the RMS roughness and the peak-valley height values of each sample (Figure 6). Indeed, peak-valley heights across assembly conditions are not statistically different (p = 0.237). However, the RMS roughness values are statistically different

ACS Paragon Plus Environment

22

Page 23 of 40

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

Biomacromolecules

and there is a real effect of assembly pH on RMS roughness (p = 0.020). Upon further analysis, Tukey’s HSD test reveals that the means between pH = 4.0 and pH = 8.0 are greater than the expected standard error. T-tests of pH = 8.0 with each of the other assembly conditions reveals statistically significant differences (p < 0.05) between pH = 4.0 and pH = 6.0. No other differences could be ascertained at a 95% confidence interval, indicating that the RMS roughness is independent of assembly pH at neutral or acidic conditions.

Figure 6. RMS roughness and peak-valley height of BMP2-(PMAA/PLH)5 multilayers as a function of assembly pH. At least three samples were measured per assembly pH.

ACS Paragon Plus Environment

23

Biomacromolecules

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 40

These results are in contrast to the work of Gong as well as Niepel et al., which indicated that the assembly pH had an impact on the PEM morphology and roughness.64,65 However, the PEM structures they considered differ from the one reported in this study in that an identical initial layer for all pH conditions was not adsorbed before building up their PEM on top of the substrate. It should be noted that AFM was performed on PEMs assembled on anodized titanium plate. Kolasinska et al. reported that the deposition of an initial layer of polyethyleneimine (PEI) could deeply impact the entire PEM structure.67 These effects included a more uniform growth of the film visible even for the first few layers, whereas layer-by-layer buildup in absence of this first PEI layer resulted in voids in the adsorbed layers. Additionally, the influence of the PEI layer could be witnessed even at the PEM surface. Films including this initial PEI layer presented a smoother surface as well as different wettability as compared to films without an initial layer of PEI, indicating that the PEI influence extended through 14 polyelectrolyte layers. Similarly, Peterson et al. reported that the addition of an initial BMP-2 layer significantly changed the surface roughness of a PEM coating.17,40 The AFM data showing no pH dependence can be explained as the competitive action of two different phenomena affecting the surface morphology. The first one is the influence of pH on the electrostatic interactions taking place between the polymer chains during the PEM formation, which modifies parameters such as the layers thickness, their entanglement and eventually the film surface roughness. The second one is the extended influence of the first layer through the whole construct. The first layer is identical for all pH conditions. Based on the AFM results displayed in Figure 6, this second phenomenon is the dominant one.

ACS Paragon Plus Environment

24

Page 25 of 40

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

Biomacromolecules

Contact Angle Analysis. In order to determine how assembly pH affected the surface energy of the resulting coating, static contact angle measurements were performed. Surface energy results are shown in Figure 7 and contact angle data are provided in the Supporting Information. The polar contribution to the surface energy was dominant for PEMs built at more acidic pH values and decreased as the basicity of the assembly pH increased, whereas the dispersive component remained effectively constant. The transition from a polar-dominated to dispersivedominated regime occurred around pH = 6.0, for which the polar and dispersive energies presented nearly identical values.

Figure 7. Polar (γsp) and dispersive (γsd) components as well as total surface energy (γs) are reported for each assembly pH. Surface energies were determined from sessile drop experiments with multiple liquids of known polar and dispersive energy. At least ten contact angles were measured per solvent. Since the final layer deposited on each PEM was PLH, the observed differences in surface energy result from the combined effects of several physical phenomena. The main one is the pH dependence of the charges present on weak polyelectrolyte chains. As the pH increases, polyanionic chains increase in charge, whereas polycationic chains will lose charge. Since the

ACS Paragon Plus Environment

25

Biomacromolecules

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 40

final PEM layer is the positively charged PLH, the charge density on the PEM surface decreases when prepared at more basic pH values and will become more and more hydrophobic, eventually leading to the observed experimental results. This is consistent with pKa values of PMAA and PLH (6.8 and 4.8/11.6, respectively).41 Above the pKa of PMAA, PMAA becomes rapidly dissociated, which may explain the sharp decrease in the polar component of surface energy. Another significant physical effect to take into account is layer interpenetration. It has been shown that the mutual interactions between polyanions and polycations in a PEM drastically change their pKa and can have a great impact on the multilayer assembly and stability, leading to the diffusion of charged chains into an oppositely charged layer.68,69 Due to this interpenetration and the generally very small thickness of these systems, the influence of buried layers may still be observed at the PEM surface and impact contact angle measurements. Additionally, the deposition pH-dependent differences in the PEM internal structure also influence the layer entanglement. Such entanglement behaviors, influence of lower layers and differences in PEM internal structure could further explain the variable surface energy observed in this study. Finally, it is interesting to note that at pH = 8.0 this system presents a surface free energy of 40.4 mJ/m² with a polar and dispersive contributions of 18.6 mJ/m² and 21.8 mJ/m². These values are close to reported values for the solid surface energy of PMAA (total : 41.0 mJ/m² , polar : 10.3 mJ/m², dispersive : 29.7 mJ/m² ).70 These results seem to hint to the system surface being energetically similar to that of solid PMAA, even though PLH was the last layer deposited. At higher pH values, PLH chains will become less charged, leading to a more hydrophobic surface, whereas the PMAA chains will gain more and more charges. Due to layer entanglement, the electrostatic influence of the lower PMAA layer on the PEM surface energy will therefore

ACS Paragon Plus Environment

26

Page 27 of 40

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

Biomacromolecules

become dominant as the PLH layer influence will decrease, eventually reaching values close to that of pure PMAA.

CONCLUSIONS BMP-2-eluting PEMs systems have received much interest from the scientific community in recent years as tailorable coatings designed to enhance the osseointegration of implants. However, the application of these systems is still hindered by a lack of understanding of their internal structure, controlled release kinetics, stimulus-dependent behavior, and how all of these parameters are linked. The work presented in this report aimed to investigate a PEM formed by an anchoring layer of BMP-2 covered by five PMAA/PLH bilayers, with the goal of determining how assembly pH impacts its structure and properties. Different BMP-2 release profiles were obtained for each pH condition. The initial amount of BMP-2 loaded was determined to be approximately 2 µg/cm², and all pH conditions exhibited a sustained release profile for at least twenty-five days. Further studies of PEMs assembled at pH = 4.0 and pH = 5.0 showed that more than two months of release could be achieved. The release profiles were not in agreement with a Fickian model, but rather with a power law model. This indicated that the release process was not relying solely on pure diffusion, and probably also involved a mass transport phenomenon due to the dynamic swelling of the PEM. The BMP-2 that was released maintained bioactivity, as determined by its ability to induce osteoblastic differentiation of myoblasts. The decrease in surface energy as the assembly pH became more basic confirmed that the changes in the internal structure also impacted the PEM surface properties. The modified layer entanglement and influence of underlying layers reflected a shift in the balance of electrostatic

ACS Paragon Plus Environment

27

Biomacromolecules

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 40

interactions caused by the assembly pH. However, the surface roughness and peak-valley height remained unchanged at different assembly pH values, indicating that the influence of the first anchoring layer was dominant over other interactions that may impact the surface topography. Future work will focus on continued characterization of this system, including quartz crystal microbalance with dissipation monitoring (QCM-D) studies combined with dynamic light scattering (DLS) results to better understand the assembly process. The role of rinsing solution conditions on PEM assembly, in particular the effect of ionic strength through addition of a buffer, is an important area of future work. Additionally, the impact of the first layer on surface properties is the focus of ongoing studies. As this BMP-2-eluting coating is designed for biomedical applications, the biological response to these surfaces will be evaluated, with the goal of disentangling the relative impacts of the physical and biochemical cues in order to design better surfaces. These data, coupled with the ones in this report and our prior work, could eventually allow for a tunable implant coating to be precisely designed and controlled in order to efficiently promote osseointegration and improve current prosthetic technologies.

ASSOCIATED CONTENT Supporting Information. Further information is available including quantification of BMP-2 left on titanium surfaces after 25 days, AFM micrographs, and static contact angle measurements. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author * Email: [email protected]; Phone: (508) 831-6029

ACS Paragon Plus Environment

28

Page 29 of 40

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

Biomacromolecules

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes The authors declare no competing financial interests. ACKNOWLEDGMENT The authors are particularly grateful for Prof. George Pins and Meagan Carnes for providing C2C12 myoblasts and for the use of the Pins lab cell culture facilities. The authors are also grateful to Prof. Nancy Burnham for use of the AFM, Prof. Chris Lambert for use of the contact angle goniometer, Prof. Dave Adams for use of the incubator, and Prof. Anjana Jain for use of the sonicator. REFERENCES (1)

U.S. Department of Health http://hcupnet.ahrq.gov/HCUPnet.jsp.

and

Human

Services.

HCUPnet

(2)

Fingar, K. R.; Stocks, C.; Weiss, A. J.; Steiner, C. A. Most Frequent Operating Room Procedures Performed in U.S. Hospitals, 2003-2012; 2014.

(3)

Weiss, A. J.; Elixhauser, A.; Andrews, R. M. Characteristics of Operating Room Procedures in U.S. Hospitals, 2011; 2014.

(4)

Bauer, T. W.; Schils, J. Skeletal Radiol. 1999, 28, 483–497.

(5)

Goodman, S. B.; Yao, Z.; Keeney, M.; Yang, F. Biomaterials 2013, 34 (13), 3174–3183.

(6)

Hu, X.; Neoh, K.-G.; Shi, Z.; Kang, E.-T.; Poh, C.; Wang, W. Biomaterials 2010, 31 (34), 8854–8863.

(7)

Macdonald, M. L.; Rodriguez, N. M.; Shah, N. J.; Hammond, P. T. Biomacromolecules 2010, 11 (8), 2053–2059.

(8)

Vrana, N. E.; Erdemli, O.; Francius, G.; Fahs, A.; Rabineau, M.; Debry, C.; Tezcaner, A.; Keskin, D.; Lavalle, P. J. Mater. Chem. B 2014, 2, 999–1008.

(9)

Shah, N. J.; Hyder, M. N.; Quadir, M. a; Dorval Courchesne, N.-M.; Seeherman, H. J.; Nevins, M.; Spector, M.; Hammond, P. T. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (35),

ACS Paragon Plus Environment

29

Biomacromolecules

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 40

12847–12852. (10)

Carragee, E. J.; Hurwitz, E. L.; Weiner, B. K. Spine J. 2011, 11 (6), 471–491.

(11)

Bessa, P. C.; Balmayor, E. R.; Hartinger, J.; Zanoni, G.; Dopler, D.; Meinl, A.; Banerjee, A.; Casal, M.; Redl, H.; Reis, R. L.; Griensven, M. Van. Tissue Eng. Part C 2010, 16 (5), 937–945.

(12)

Nath, S. D.; Abueva, C.; Kim, B.; Lee, B. T. Carbohydr. Polym. 2015, 115, 160–169.

(13)

Crouzier, T.; Sailhan, F.; Becquart, P.; Guillot, R.; Logeart-Avramoglou, D.; Picart, C. Biomaterials 2011, 32 (30), 7543–7554.

(14)

Macdonald, M. L.; Samuel, R. E.; Shah, N. J.; Padera, R. F.; Beben, Y. M.; Hammond, P. T. Biomaterials 2011, 32 (5), 1446–1453.

(15)

Cai, P.; Xue, Z.; Qi, W.; Wang, H. Colloids Surfaces A Physicochem. Eng. Asp. 2013, 434, 110–117.

(16)

Gand, A.; Hindié, M.; Chacon, D.; van Tassel, P. R.; Pauthe, E. Biomatter 2014, 4 (February), e28823.

(17)

Peterson, A. M.; Pilz-Allen, C.; Kolesnikova, T.; Möhwald, H.; Shchukin, D. G. ACS Appl. Mater. Interfaces 2014, 6 (3), 1866–1871.

(18)

Abbah, S.-A.; Liu, J.; Lam, R. W. M.; Goh, J. C. H.; Wong, H.-K. J. Control. Release 2012, 162 (2), 364–372.

(19)

Meyer, F.; Dimitrova, M.; Jedrzejenska, J.; Arntz, Y.; Schaaf, P.; Frisch, B.; Voegel, J. C.; Ogier, J. Biomaterials 2008, 29, 618–624.

(20)

Lee, I.-C.; Wu, Y.-C. ACS Appl. Mater. Interfaces 2014, 6, 14439–14450.

(21)

Zhang, C.; Hirt, D. E. Polymer (Guildf). 2007, 48 (23), 6748–6754.

(22)

Huang, X.; Zacharia, N. S. Soft Matter 2013, 9, 7735.

(23)

Fakhrullin, R. F.; Zamaleeva, A. I.; Minullina, R. T.; Konnova, S. A.; Paunov, V. N. Chem. Soc. Rev. 2012, 41 (11), 4189–4206.

(24)

Zhang, J.; Chua, L. S.; Lynn, D. M. Langmuir 2004, 20 (24), 8015–8021.

(25)

Moskowitz, J. S.; Blaisse, M. R.; Samuel, R. E.; Hsu, H. P.; Harris, M. B.; Martin, S. D.; Lee, J. C.; Spector, M.; Hammond, P. T. Biomaterials 2010, 31 (23), 6019–6030.

(26)

Shukla, A.; Fuller, R. C.; Hammond, P. T. J. Control. Release 2011, 155 (2), 159–166.

(27)

Shukla, A.; Fleming, K. E.; Chuang, H. F.; Chau, T. M.; Loose, C. R.; Stephanopoulos, G. N.; Hammond, P. T. Biomaterials 2010, 31 (8), 2348–2357.

(28)

Guillot, R.; Gilde, F.; Becquart, P.; Sailhan, F.; Lapeyrere, A.; Logeart-Avramoglou, D.; Picart, C. Biomaterials 2013, 34 (23), 5737–5746.

(29)

Wu, H.; Wang, S.; Fang, H.; Zan, X.; Zhang, J.; Wan, Y. Colloids Surf. B. Biointerfaces 2011, 82 (2), 602–608.

(30)

Shah, N. J.; Macdonald, M. L.; Beben, Y. M.; Padera, R. F.; Samuel, R. E.; Hammond, P. T. Biomaterials 2011, 32 (26), 6183–6193.

(31)

Gilde, F.; Maniti, O.; Guillot, R.; Mano, J. F.; Logeart-Avramoglou, D.; Sailhan, F.;

ACS Paragon Plus Environment

30

Page 31 of 40

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

Biomacromolecules

Picart, C. Biomacromolecules 2012, 13 (11), 3620–3626. (32)

Schönhoff, M.; Ball, V.; Bausch, A. R.; Dejugnat, C.; Delorme, N.; Glinel, K.; von Klitzing, R.; Steitz, R. Colloids Surfaces A Physicochem. Eng. Asp. 2007, 303, 14–29.

(33)

Sukhishvili, S. A. Curr. Opin. Colloid Interface Sci. 2005, 10 (1-2), 37–44.

(34)

Skirtach, A. G.; Yashchenok, A. M.; Möhwald, H. Chem. Commun. 2011, 47, 12736– 12746.

(35)

Glinel, K.; Déjugnat, C.; Prevot, M.; Schöler, B.; Schönhoff, M.; von Klitzing, R. Colloids Surfaces A Physicochem. Eng. Asp. 2007, 303, 3–13.

(36)

Sato, K.; Yoshida, K.; Takahashi, S.; Anzai, J. Adv. Drug Deliv. Rev. 2011, 63 (9), 809– 821.

(37)

Pavlukhina, S.; Sukhishvili, S. Adv. Drug Deliv. Rev. 2011, 63 (9), 822–836.

(38)

Huang, Y.; Luo, Q.; Li, X.; Zhang, F.; Zhao, S. Acta Biomater. 2012, 8 (2), 866–877.

(39)

Thompson, M. T.; Berg, M. C.; Tobias, I. S.; Rubner, M. F.; Van Vliet, K. Biomaterials 2005, 26, 6836–6845.

(40)

Peterson, A. M.; Pilz-Allen, C.; Möhwald, H.; Shchukin, D. G. J. Mater. Chem. B 2014, 2 (18), 2680–2687.

(41)

Peterson, A. M.; Möhwald, H.; Shchukin, D. G. Biomacromolecules 2012, 13 (10), 3120– 3126.

(42)

Ishida, N.; Biggs, S. Langmuir 2007, 23 (22), 11083–11088.

(43)

Vidyasagar, A.; Sung, C.; Gamble, R.; Lutkenhaus, J. L. ACS Nano 2012, 6 (7), 6174– 6184.

(44)

Vidyasagar, A.; Sung, C.; Losensky, K.; Lutkenhaus, J. L. Macromolecules 2012, 45 (22), 9169–9176.

(45)

Katagiri, T.; Yamaguchi, A.; Komaki, M.; Abe, E.; Takahashi, N.; Ikeda, T.; Rosen, V.; Wozney, J. M.; Fujisawa-Sehara, A.; Suda, T. J. Cell Biol. 1994, 127 (6 I), 1755–1766.

(46)

Crouzier, T.; Ren, K.; Nicolas, C.; Roy, C.; Picart, C. Small 2009, 5 (5), 598–608.

(47)

Grasman, J. M.; Do, D. M.; Page, R. L.; Pins, G. D. Biomaterials 2015, 72, 49–60.

(48)

Hughes-Fulford, M.; Li, C.-F. J. Orthop. Surg. Res. 2011, 6, 1–8.

(49)

Owens, D. K.; Wendt, R. C. J. Appl. Polym. Sci. 1969, 13 (8), 1741–1747.

(50)

Mauser, T.; Déjugnat, C.; Möhwald, H.; Sukhorukov, G. B. Langmuir 2006, 22 (13), 5888–5893.

(51)

Antipov, A. a.; Sukhorukov, G. B.; Leporatti, S.; Radtchenko, I. L.; Donath, E.; Möhwald, H. Colloids Surfaces A Physicochem. Eng. Asp. 2002, 198-200, 535–541.

(52)

Antipov, A. A.; Sukhorukov, G. B. Adv. Colloid Interface Sci. 2004, 111 (1-2), 49–61.

(53)

Chug, A.; Shukla, S.; Mahesh, L.; Jadwani, S. J. Oral Maxillofac. Surgery, Med. Pathol. 2013, 25 (1), 1–4.

(54)

Pegg, E.; Mellon, S.; Gill, H. In Bone-Implant Interface in Orthopedic Surgery SE - 2; Karachalios, T., Ed.; Springer London, 2014; pp 13–26.

ACS Paragon Plus Environment

31

Biomacromolecules

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 40

(55)

Huang, X.; Brazel, C. S. J. Control. Release 2001, 73 (2-3), 121–136.

(56)

Siepmann, J.; Peppas, N. A. Adv. Drug Deliv. Rev. 2001, 48, 139–157.

(57)

Siepmann, J.; Peppas, N. A. Int. J. Pharm. 2011, 418 (1), 6–12.

(58)

Liu, T. Y.; Lin, Y. L. Acta Biomater. 2010, 6 (4), 1423–1429.

(59)

Magboo, R.; Peterson, A. M. Encyclopedia of Surface and Colloid Science, Third Edition; 2014.

(60)

Berg, M. C.; Zhai, L.; Cohen, R. E.; Rubner, M. F. Biomacromolecules 2006, 7, 357–364.

(61)

Varshosaz, J.; Falamarzian, M. Eur. J. Pharm. Biopharm. 2001, 51 (3), 235–240.

(62)

Samuel, R. E.; Shukla, A.; Paik, D. H.; Wang, M. X.; Fang, J. C.; Schmidt, D. J.; Hammond, P. T. Biomaterials 2011, 32 (30), 7491–7502.

(63)

Mhamdi, L.; Picart, C.; Lagneau, C.; Othmane, A.; Grosgogeat, B.; Jaffrezic-Renault, N.; Ponsonnet, L. Mater. Sci. Eng. C 2006, 26, 273–281.

(64)

Gong, X. Phys. Chem. Chem. Phys. 2013, 15, 10459–10465.

(65)

Niepel, M. S.; Peschel, D.; Sisquella, X.; Planell, J. A.; Groth, T. Biomaterials 2009, 30 (28), 4939–4947.

(66)

Quinn, J. F.; Caruso, F. Langmuir 2004, 20 (13), 20–22.

(67)

Kolasińska, M.; Krastev, R.; Warszyński, P. J. Colloid Interface Sci. 2007, 305 (1), 46– 56.

(68)

Burke, S. E.; Barrett, C. J. Langmuir 2006, 19 (22), 3297–3303.

(69)

Itano, K.; Choi, J.; Rubner, M. F. Macromolecules 2005, 38, 3450–3460.

(70)

Solid surface energy (SFE) for common polymers http://www.surface-tension.de/solidsurface-energy.htm.

ACS Paragon Plus Environment

32

Page 33 of 40

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

Biomacromolecules

Table of Contents Graphic 87x34mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

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

Figure 1. Cumulative released BMP-2 from PEMs with various assembly pH values. Error bars represent standard deviation. 83x65mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 34 of 40

Page 35 of 40

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

Biomacromolecules

Figure 2. Cumulative percentage of released BMP-2 for PEMs assembled at pH=4.0 and pH=5.0 in terms of a. mass and b. percentage. Error bars represent standard deviation. The total amount of BMP-2 released at the end of the experiment was identified to be the total amount of BMP-2 initially adsorbed. 83x130mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

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

Cumulative BMP-2 released from an uncoated anodized titanium surface. Error bars represent standard deviation, n = 4. 83x64mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 36 of 40

Page 37 of 40

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

Biomacromolecules

(PH/PMAA)5.5 mass under release study conditions. a. PEM mass; b. Change in PEM mass. (PH/PMAA)5.5 PEMs were assembled at pH = 4.0, 6.0 or 8.0 on titanium sensors. Release study conditions were 37 °C in a PBS solution. Error bars represent standard deviation, n = 4. 167x64mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

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

Alkaline phosphatase enzyme activity of C2C12 cells cultured on different surfaces. Ti is an anodized titanium surface; Ti +BMP-2 is an anodized titanium surface, 2 µg BMP-2 was added to the growth medium at time t = 0; pH = 4 is a (PH/PMAA)5.5 PEM assembled on anodized titanium at pH = 4.0; pH = 4 + BMP-2 is BMP-2-(PMAA/PH)5 assembled on anodized titanium at pH = 4.0. Error bars represent standard deviation, n = 3. 83x64mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 38 of 40

Page 39 of 40

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

Biomacromolecules

RMS roughness and peak-valley height of BMP2-(PMAA/PLH)5 multilayers as a function of assembly pH. At least three samples were measured per assembly pH. 83x130mm (300 x 300 DPI)

ACS Paragon Plus Environment

Biomacromolecules

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

Polar (γsp) and dispersive (γsd) components as well as total surface energy (γs) are reported for each assembly pH. Surface energies were determined from sessile drop experiments with multiple liquids of known polar and dispersive energy. At least ten contact angles were measured per solvent. 83x65mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 40 of 40