Enzymatic Degradation of Polysaccharide-Based ... - ACS Publications

Subscriber access provided by RMIT University Library

Article

Enzymatic degradation of polysaccharide-based layer-by-layer structures Matias J Cardoso, Sofia G Caridade, Rui R. Costa, and João F. Mano Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.5b01742 • Publication Date (Web): 08 Mar 2016 Downloaded from http://pubs.acs.org on March 10, 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 39

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

Enzymatic degradation of polysaccharide-based layer-by-layer structures Matias J. Cardoso†,‡, Sofia G. Caridade†,‡, Rui R. Costa†,‡, João F. Mano†,‡,* †3B’s Research Group – Biomaterials, Biodegradables and Biomimetics, University of Minho, Headquarters of the European Institute of Excellence of Tissue Engineering and Regenerative Medicine, Avepark – Parque de Ciência e Tecnologia, Zona Industrial da Gandra, 4805-017 Barco GMR, Portugal. ‡ICVS/3B’s, PT Government Associated Laboratory, Braga/Guimarães, Portugal. KEYWORDS. Layer-by-layer; Ultrathin films; Freestanding films; Microcapsules; Enzymatic degradation; Biomaterials; Drug release.

ABSTRACT. The lack of knowledge on the degradation of layer-by-layer structures is one of the causes hindering its translation to preclinical assays. The enzymatic degradation of chitosan/hyaluronic acid films in the form of ultrathin films, freestanding membranes, and microcapsules was studied resorting to hyaluronidase. The reduction of the thickness of ultrathin films was dependent on the hyaluronidase concentration, leading to thickness and topography 1

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

variations. Freestanding membranes exhibited accelerated weight loss up to 120 hours in the presence of the enzyme, achieving complete degradation. Microcapsules with around 5 µm loaded simultaneously with FITC-BSA and hyaluronidase showed that the co-encapsulation of such enzyme and protein mixture led to a FITC-BSA release four times higher than in the absence of hyaluronidase. The results suggest that the degradation of LbL devices may be tuned via embedded enzymes, namely in the controlled release of active agents in biomedical applications.

INTRODUCTION The performance of polymeric and composite structures such as drug delivery systems and implantable devices (including scaffolds and hydrogels for tissue engineering) is often dependent on their stability and integrity when introduced in biological environments.1-3 Polymeric structures can be susceptible to various types of degradation, including thermal, photo, mechanical and chemical mechanisms.4 In the case of structures based on natural polymers, enzyme-driven degradation plays a major role.5 In tissue engineering, for example, polymeric scaffolds aim to degrade ideally at the rhythm of the target tissue regeneration while promoting colonization by native cells.6-8 Drug carriers for targeted therapies should retain and protect a drug while in circulation in the bloodstream until cellular uptake and intracellular degradation.911

Regardless of the envisaged application, one should predict the physical, chemical and

structural changes on biomedical devices induced by enzymes of the biological environment. 2


Page 2 of 39

Page 3 of 39

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

It is clear that evaluating the biodegradability of biomedical devices is crucial not only for current medical technologies but also for new cutting-edge tissue engineering ones. It is the case of layer-by-layer (LbL) devices, which are obtained by a simple process of adsorption of two or more biomaterials to a substrate in a sequential fashion.12 Such adsorption requires the use of materials that are capable of interacting via complemental interactions (e.g. electrostatic interactions, hydrogen bonds, hydrophobic interactions) to generate nanostructured and multilayered structures exhibiting thickness ranging from several nanometers to a few micrometers.13-15 LbL is a versatile technique that can be used to fabricate numerous structures such as planar ultrathin films and membranes, as well as three-dimensional tubular/spherical shapes and porous scaffolds, without resorting to aggressive solvents. Due to the variety of geometries and scaling potential, LbL devices have been suggested for several biomedical applications, including biosensors, drug delivery, coating of biomaterials, and tissue engineering.16 However, the biodegradability of LbL devices has been seldom explored,17-22 which hinders its transition to translational and preclinical studies.23,24 For example, Etienne et al. studied the enzymatic degradation of multilayer films in vitro and in vivo, the latter in the oral environment.25 The degradation of microcapsules was also studied by Szarpak et al., who analyzed the morphology, permeability and enzymatic degradation of microcapsules.26 In this work, we will explore the susceptibility of nano- and micro-sized LbL structures based on natural polysaccharides to enzymatic degradation. Natural materials have raised much interest as LbL ingredients. Unlike synthetic materials, natural materials are often biocompatibility, noncytotoxic, biodegradable and exhibit native bioactivity.7,27,28 Among them, polysaccharides are especially interesting, as they can be highly hydrated, biocompatible, biodegradable 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

specifically by enzymatic action, and can be chemically modified and cross-linked to adjust the degradation rate required for a specific application.29 Furthermore, as many polysaccharides exhibit charged groups, they can be easily processed into polyelectrolyte multilayers exploiting electrostatic interactions. Chitosan (CHT) and hyaluronic acid (HA) are two abundant natural polysaccharides exhibiting positive and negative charge, respectively.30-32 Besides their bioavailability and electrostatic complementarity, CHT and HA are common LbL building blocks, thus being good candidates to integrate future LbL-based products.21,33,34 The biodegradability of such devices should take into consideration the existence of enzymes in the human body. For example, hyaluronidase (Hase) is an enzyme that degrades HA. In mammalians, Hases are classified in different groups depending on the cleavage site on the polymer chain, which can be random or specific (e.g. endo-hexosaminidases).35,36 In the human body, Hase is found in different organs, like in skin, eye, liver, kidney, uterus and placenta, as well as in various physiological fluids.37 In this work, the enzymatic degradability of CHT/HA ultrathin films, freestanding membranes (also known as freestandings) and microcapsules will be addressed using Hase. The influence of Hase on the stability of such structures will be compared by means of thickness and topography variations in ultrathin films. The results will be extrapolated to freestandings and evaluated in terms of weight loss and variations of mechanical properties. We will also explore the susceptibility of the enzymatic degradation of CHT/HA multilayers to force and control the release of active agents from compartmentalized drug release systems. The use of preencapsulated Hase to control the release profile of drugs from LbL microcarriers will be herein introduced based on the quantification of FITC-BSA release. 4


Page 4 of 39

Page 5 of 39

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

EXPERIMENTAL SECTION Materials. Medium molecular weight chitosan with 75-85% degree of deacetylation (DD) (ref 448877), hyaluronic acid sodium salt from Streptococcus equi (ref 53747), hyaluronidase from bovine testes (isoelectric point≈5.4),38 Type I-S, lyophilized powder, 400-1000 units/mg solid (ref H3506), albumin fluorescein isothiocyanate conjugate (ref A9771), phosphate buffered saline tablets (PBS, ref P4417), sodium carbonate ACS reagent anhydrous (ref 222321), and ethylenediaminetetraacetic acid (EDTA, ref E9884) were purchased from Sigma-Aldrich. Sodium chloride (ref 131659) was purchased from Laborspirit (Portugal). Calcium chloride (ref 1.02378) and glutaraldehyde 25% (ref 1.04239.0250) were purchased from VWR international (Portugal). Build-up and enzymatic degradation of thin films by quartz-crystal microbalance. A quartzcrystal microbalance with dissipation monitoring (QCM-D) (Q-Sense, E4 model, Sweden) and gold-coated AT-cut quartz sensors (Q-sense, ref. QSX 301) was used to follow the adsorption of CHT and HA, with simultaneous excitation of multiple overtones: 1st, 3rd, 5th, 7th, 9th, 11th, and 13th, corresponding to: 5, 15, 25, 35, 45, 55, and 65 MHz, respectively. Adsorption took place at 25 ºC using polyelectrolyte solutions prepared at 0.5 mg·mL-1 in NaCl 0.15 M, pH=5.5, intercalated with a rinsing step with NaCl 0.15 M. All deposition and rinsing solutions were flushed for 10 min at a constant flow rate of 50 µL·min-1. Monitoring of the LbL assembly proceeded until three bilayers of CHI/ALG – (CHT/HA)3 – or three bilayers with an extra layer of CHT – (CHI/HA)3-CHI – were assembled. After reaching the desired number of layers, the 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 39

temperature was raised to 37 ºC (heating slope: 1 ºC per minute). Then, the Hase solution at different concentrations (1, 10, 50, 100, 1000 and 2000 µg·mL-1; pH=5.5) was flushed for 12 min at a constant flow rate of 25 µL·min-1, after which the flow was stopped and Hase was left to incubate for 24h. After 24h, a rinsing step time of 20 min followed to remove remains of Hase and film debris. The registered frequencies for each overtone (∆fn) are already normalized to the fundamental resonant frequency of the quartz crystal. To estimate the film thickness, the Voigt-based model was used,39 contained in the software QTools (version: 3.1.25.604) provided by Q-Sense, based on Equations 1 and 2, 2   η jω 2 η 3  η3     ∆F ≈ − + h − 2 h ρ ω  ∑ j j j 2 2   2 2πρ 0 h0 δ 3 j = k  δ 3  µ j + ω η j  

(1)

2   µ jω η 3  η3    [ ∆D ≈ + 2 h  ∑ j  δ  µ 2 + ω 2η 2   2πfρ 0 h0 δ 3 j =k  3 j j  

(2)

1

[

1

where, considering a total of k thin viscoelastic layers, ρ0 and h0 are the density and thickness of the quartz crystal, η3 is the viscosity of the bulk liquid, δ3 is the viscous penetration depth of the shear wave in the bulk liquid, ρ3 is the density of liquid, µ is the elastic shear modulus of an overlayer, and ω is the angular frequency of the oscillation. The model requires three parameters, namely solvent density, solvent viscosity and film density, to be estimated. The solvent viscosity was therefore fixed at 0.001 Pa (the same as for water) and film density at 1200 kg·m-3. The solvent density was varied by trial and error between 1000 and 1015 kg.m-3 until the total error, χ2, was minimized. Calculation were made using three overtones (5th, 7th, 9th). The area density was calculated by multiplying the thickness by the film density.

6


Page 7 of 39

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

Construction and enzymatic degradation of thin films by atomic force microscopy. Atomic force microscopy (AFM, Dimension Icon, Brunker, France) was used to analyze the effect of Hase on the (CHT/HA)3 ultrathin films surfaces. The analyzed films were built by dipping microscopy glass slides (1×1 cm2) alternately in the polyelectrolyte solutions. The glass slides had been previously cleaned with acetone, ethanol, 2-propanol and ultrapure water, and dried with nitrogen gas. Adsorption took place at room temperature, using the same conditions of QCM-D studies. Then, Hase solution at 50 µg·mL-1 was added (3 mL) and the samples were incubated at 37 ºC during pre-established periods (3, 6, 9 and 24h). Thereafter, the samples were rinsed with ultrapure water to remove enzyme and salt remains, and kept in ultrapure water until AFM analysis (no more than 24h). Prior to mounting the sample in the AFM sample holder, excess water was removed with the aid of a paper filter placed at the corner of the sample. Thin films topographies were acquired at room temperature with a 512×512 pixel resolution, analyzing areas of 5×5 µm2 to calculate the arithmetic mean and root mean squared roughness (Ra and Rq, respectively). At least three measurements were performed on replica specimens (n=3). Production of chitosan-hyaluronic acid freestanding films by dipping robot. CHT/HA freestanding membranes were produced using an in-house automatized robot designed for fabrication of multilayers membranes. CHT and HA at 2 mg·mL-1, NaCl 0.15 M, pH=5.5 were used to obtain robust and detachable films. The multilayer films were fabricated on a polypropylene substrate that promotes an easily detachable membrane.40,41 Polypropylene substrates were immersed alternately in CHT and HA solutions for 6 min with an intermediate

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 39

rinsing step in NaCl 0.15 M, pH=5.5 for 4 min. This cycle was repeated until 100 bilayers were adsorbed, after which the membranes were dried at room temperature and stored until use. Enzymatic degradation tests. (CHT/HA)100 membranes (20×20 mm2) were mounted in a polypropylene holder designed especially for the experiment. The polypropylene holder (80×30 mm2) has a frame in the middle (13×13 mm2) to promote contact between the membrane and the enzymatic solution, and also to prevent self-folding and stickiness against the container of the enzyme solution. All membranes were first weighted (initial mass, mi) and then mounted. The membrane holder was placed in a cylindrical tube containing 40 mL of Hase at 50 µg·mL-1 prepared in NaCl 0.15 M, pH=5.5. Then, the holders were placed at 37 ºC and retrieved after predetermined time-points: 3, 6, 9 and 24h. Control samples were placed in tubes containing only NaCl 0.15 M, pH=5.5. Following the retrieval, the holders were meticulously rinsed with ultrapure water, followed by drying at room temperature. Dry membranes were detached from the holder and weighted (final mass, mf). Weight loss was calculated using Equation 3, Weight loss (%) =

mi − m f mi

× 100

(3)

For the study of long-term degradation, membranes with the same dimensions were mounted in the holders and immersed in 40 mL of Hase at 50 µg·mL-1 prepared in PBS (pH=7.4). Then, the holders with the samples were placed at 37 ºC and retrieved after predetermined time-points: 6, 9, 18, 24, 48, 72 and 120h. Following the retrieval, the holder were meticulously rinsed with ultrapure water, followed by drying at room temperature. Dry membranes were detached from the holder and weighted. Equation 3 was once again used to calculate the weight loss. All these procedures were made in triplicate (n=3). 8


Page 9 of 39

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

Real-time degradation monitoring by Dynamic Mechanical Analysis (DMA). To promote the stability of the freestandings and allow proper handling, the freestandings were lightly crosslinked using 1% (w/v) glutaraldehyde for 1 min.42,43 The freestandings were subsequently rinsed with ultrapure water to remove glutaraldehyde remains, dried at room temperature and stored until tested. All of the viscoelastic measurements were performed using a TRITEC2000B DMA from Triton Technology (United Kingdom), equipped with tensile mode. The measurements were carried out at 37 ºC. The distance between the clamps was 5 mm and the membrane samples were cut with 10 mm width. Samples were always analyzed immersed in a liquid bath placed in a Teflon reservoir. Membranes were incubated in NaCl 0.15 M (pH=5.5) prior to their placement in the equipment. Samples were clamped in the DMA apparatus and immersed in a liquid bath. Different bath compositions were used: (i) Hase 50 µg·mL-1 prepared in NaCl 0.15 M (pH=5.5), and (ii) Hase-free NaCl 0.15 M (pH=5.5) used as control. After equilibrium at 37 ºC, the mechanical/viscoelastic properties of the samples were recorded at 1 Hz during 24h. The experiments were performed under constant strain amplitude (50 µm). A static preload of 1 N was applied during the tests to keep the sample tight. Note that the elastic modulus that is extracted during the measurement, E’, is calculated using the initial thickness of the membrane. Such value is assumed constant during the real time tests and thus E’ should be considered as an apparent elastic modulus. Entrapment of protein in calcium carbonate (CaCO3) microparticles. FITC-BSA and Hase were co-encapsulated in CaCO3 microparticles following a well-known method44-46. Shortly, to encapsulate 8 mg of FITC-BSA, aqueous solutions of sodium carbonate (Na2CO3) and calcium chloride (CaCl2) were prepared at 1 M. Co-precipitation of both solutions was performed under 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 39

vigorous stirring (≈1000 rpm) by adding 1 mL of Na2CO3 into a mixture composed of 1 mL of CaCl2 and 4 mL of FITC-BSA at 2 mg·mL-1 and Hase at 10 µg.mL-1. After 30s, stirring was stopped and the suspension of newly synthesized calcium carbonate (CaCO3) microparticles was left to react and precipitate for 15 min. The supernatant was removed and the particles washed twice with ultrapure water to remove residual salts and non-entrapped FITC-BSA and Hase. Each supernatant was retained for determination of entrapment losses. The same method was followed for CaCO3 microparticles entrapping solely FITC-BSA to be used as a control. Construction of the multilayer microcapsules. The CaCO3 sacrificial templates entrapping FITC-BSA and Hase were immersed alternately in CHT and HA solutions at 0.5 mg·mL-1 in NaCl 0.15 M (pH=5.5) for 10 min each under mild agitation, intercalated with NaCl 0.15 M (pH=5.5) during washing stages. To exchange solutions, the agitations was stopped and the particles left to precipitate, after which the aqueous medium was retrieved and replaced by the next one. This process was repeated to increase the number of bilayers until three CHT/HA bilayers – (CHT/HA)3 – were assembled. After construction, the CaCO3 core was removed by immersing the coated particles in EDTA 0.2 M pH=7.4 for 30 min, a chelating agent of Ca2+ ions. All supernatant and polyelectrolyte solutions were retained for fluorescence measurements and quantification of FITC-BSA losses during microcapsule construction. After all the steps comprising the entrapment of 8 mg of FITC-BSA, template coating and losses quantification, the produced microcapsules batch was split in three equal parts to conduct triplicate experiments. To quantify FITC-BSA release, 100% was considered to be the total mass remaining in the microcapsules after their construction.

10


Page 11 of 39

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

Determination of protein encapsulation efficiency. Fluorescence measurements were performed to establish a calibration curve and calculate FITC-BSA mass losses using microplate reader (BioTek, USA). 150 µL of sample was pipetted into a white 96-well plate in triplicate. PBS was used as blank. The excitation (λexc) and emission (λem) wavelengths used were 485 and 528 nm, respectively. The concentration range for the calibration curve was 2 to 40 µg·mL-1. All supernatants collected during entrapment, construction and chelation were measured and quantified. The encapsulation efficiency was calculated by subtracting the cumulative losses from the initial FITC-BSA mass. Quantification of FITC-BSA release. (CHT/HA)3 microcapsules loaded with (i) FITC-BSA plus Hase, or (ii) solely FITC-BSA, were resuspended in 5 mL of PBS at 37 ºC for 14 days. At each predetermined time-point (1, 2, 3, 5, 7, 9, 12 and 14 days) 450 µL was retrieved and refreshed with new PBS. The fluorescence of the retrieved samples was measured following the same quantification procedure described for the determination of protein encapsulation efficacy. All experiments were performed in triplicate (n=3). Microscopy characterization of CaCO3 microparticles and multilayer capsule shells. A scanning electron microscopy (SEM, JEOL model JSM-6010LV, Japan), was used to evaluate the morphology of the protein-loaded CaCO3 particles. Dry particles were precoated with a conductive layer of sputtered gold and then observed. Hydrated CHT/HA multilayer microcapsules loaded with FITC-BSA and Hase were observed with a TCSP8 confocal laser scanning microscope (Leica, Germany) using samples capsules suspended in a droplet of PBS. Statistical analysis. Values reported are means±1 standard deviation (SD) of at least three independent experiments. All values were analyzed using ANOVA statistical analysis using 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 39

Microsoft Excel (version 15.0.4771.1000, Microsoft, USA). All results were considered to be statistically significant at p-value less than 0.05 (*), 0.01 (**), or 0.001 (***).

RESULTS AND DISCUSSION Enzymatic degradation of chitosan/hyaluronic acid ultrathin films. In order to study the enzymatic degradation of CHT/HA polyelectrolyte multilayer ultrathin coatings, the construction of the film, as well as the effect of Hase, was followed in situ using QCM-D. In this technique, it is possible to follow the build-up of the films in real time by applying an alternating electric field across a gold-coated quartz crystal.47 Based on the piezoelectric effect, an alteration of mass at the surface is converted to variations of frequency (∆f). Monitoring the dissipation also allows to measure the damping properties (i.e. viscoelasticity) of the film (∆D). A major advantage of this technique is its capability to detect changes induced by elements found in biological environments, such as the presence of enzymes. Figure 1A shows the variations of frequency (∆f5/5) and dissipation (∆D5) for the 5th overtone during the construction of 3 bilayers of CHT/HA films and one last layer of CHT – henceforth referred to as (CHT/HA)3-CHT. The role of the last layer of CHT was to determine how different end-layer polymers may affect the degradation behavior of a multilayer film. As time progressed, the sequential decrease of ∆f5/5 values to an approximate value of -375 Hz shows a successful film construction, evidencing the deposition of polymer molecules onto the surface. An increase of dissipation was also registered, revealing that the films were not rigid. Furthermore, such variation represents a shift towards a film with a higher viscous component and great ability to absorb water, properties that are typical of soft polymeric structures. 12


Page 13 of 39

0

∆f5/5

-50

60

∆D5

∆f5/5 (Hz)

-100 -150

50

1 2 3 4

40

-200 30

-250 -300

20

-350

10

∆D5 (10-6)

(A)

-400 0 0

1

2

Time (h) (B) 80

Thickness (nm)

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

60

40

20

0 0

1

2

3

4

5

6

7

Layers

Figure 1. (A) Representative QCM-D results of in situ build-up of (CHT/HA)3-CHT thin films for the 5th overtone with normalized variations of frequency (∆f5/5, ●), and dissipation (∆D5, □). (1) Addition of CHT; (2) and (4) rinsing steps with NaCl 0.15 M; (3) addition of HA. (B) Representative cumulative thickness variations of (CHT/HA)3-CHT films.

The QCM-D data was used to determinate the film thickness in every adsorption step resorting to the Voigt-based viscoelastic model. Figure 1B shows the cumulative thickness evolution of a (CHT/HA)3-CHT film, which reached a value of 79 nm. (CHT/HA)3 films, i.e. without the end13


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 39

layer of CHT, reached a value of 75 nm. Table 1 shows the contribution of each adsorbed layer to the thickness and area density of the films up to 7 layers.

Table 1. Thickness and area density contributions with each incremental layer. Odd layers represent CHT and even layers represent HA. Layer

Thickness

Area density

number

(nm)

(µg·cm-2)

1

4

0.48

2

6

0.72

3

10

1.2

4

8

0.96

5

12

1.44

6

12

1.44

7

27

3.24

Considering films with 6 layers, the contribution to the area density of CHT and HA are similar (i.e. a total of 3.12 µg·cm-2 each). The QCM-D experiments confirm that CHT can be used with HA to obtain ultrathin films exhibiting viscoelastic properties. It was previously reported that the thickness of CHT/HA films grows exponentially, due to the polycation ability to diffuse “in” and “out” of the whole film at each deposition step.25,33,48 The nonlinear evolution of the thickness in Figure 1B is consistent with such observations. 14


Page 15 of 39

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

After the film construction, ∆fn were recorded for further 24h at 37 ºC (to simulate body temperature) after flushing them with Hase solutions prepared at various concentrations, adjusted to pH=5.5. This pH value was selected not only to ensure optimal enzyme activity (between 4.5-6.0)49 but also to avoid inducing pH-dependent variations on the film, which would result in disruptions on the ∆fn and ∆Dn values. After 24h, the enzyme solution was removed alongside with film remains by flushing with NaCl 0.15 M for 30 min. The variations of -∆f5/5 for 24h after Hase injection and the last rinsing step evidence the various degradations profiles induced by the contact of Hase with the films and the effect of varying Hase concentration (Figure 2). A decrease of frequency in absolute value indicates removal of film mass, thus being an indicator of degradation. For (CHT/HA)3 films (Figure 2A), -∆f5/5 decreased faster as higher enzyme concentrations were used. In particular, it was observed that for the highest Hase concentrations (100, 1000 and 2000 µg·mL-1) most of the film was degraded in the first 2h. For example, -∆f5/5 decreased about 70% in 2h when enzyme concentration was 2000 µg·mL-1. Afterwards, the -∆f5/5 decrease rate was progressively slower during the remaining hours of the experiment. At the lowest concentrations, the variations were not as fast as with the highest concentrations but they were also progressively slower during the monitored 24h. At the minimum studied enzyme concentration (1 µg·mL-1), though a pronounced -∆f5/5 decrease was observed in the first hour, the variation was only residual during the following hours (about 20% of -∆f5/5 decrease in respect to the initial value). Using a Hase solution with a concentration of 50 µg·mL-1 provided an intermediate -∆f5/5 decrease rate among all the tested concentrations. In the case of (CHT/HA)3-CHT films, similar -∆f5/5 variations were observed (Figure 2B), although not as pronounced for the two lowest Hase concentrations. This may be due to the Hase 15


Biomacromolecules

specificity. Hase randomly cleaves β-N-acetylhexosamine-[1→4] glycosidic bonds in HA. Thus, films ending in CHT exhibit a more delayed degradation, since CHT is not a specific substrate of Hase.50

(CHT/HA)3 (A) 500

-1 1 µg.mL

450

-1

(CHT/HA)3-CHT (B) 500

-1 50 µg.mL

1000 µg.mL

-1

2000 µg.mL

1 µg.mL

450

-1

400

350

−∆f5/5 (Hz)

−∆f5/5 (Hz)

10 µg.mL

-1 100 µg.mL -1 0 µg.mL

400

300 250 200 150

-1

-1 10 µg.mL

-1 50 µg.mL

-1 100 µg.mL -1 0 µg.mL

-1 1000 µg.mL

4

12

-1 2000 µg.mL

350 300 250 200 150

100

100

50

50 0

0 0

4

8

12

16

20

0

24

8

16

20

24

Time (h)

Time (h) (C)

(D) 1 µg.mL

100

-1 50 µg.mL -1 100 µg.mL

-1

10 µg.mL

-1

-1 1000 µg.mL -1 2000 µg.mL

Thickness reduction (%)

Thickness reduction (%)

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 39

80 60 40 20

100

-1 50 µg.mL -1 100 µg.mL

-1 1 µg.mL -1 10 µg.mL

-1 1000 µg.mL -1 2000 µg.mL

80 60 40 20 0

0 2h

6h

12h

2h

24h

6h

12h

24h

Time

Time

Figure 2. Normalized frequency variations for the 5th overtone (-∆f5/5) during 24h for (A) (CHTHA)3 thin films and (B) (CHT-HA)3-CHT, after adding Hase at different concentrations: 16


Page 17 of 39

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

0 µg·mL-1 (►); 1 µg·mL-1 (■); 10 µg·mL-1 (●); 50 µg·mL-1 (▲); 100 µg·mL-1 (▼); 1000 µg·mL-1 (♦); 2000 µg·mL-1 (◄). Thickness reduction of (C) (CHT/HA)3, and (D) (CHT/HA)3-CHT films after 24h of Hase incubation at 37 ºC as quantified with the QCM-D data using the Voigt-based model. Thickness reduction for 0 µg·mL-1 were too low to be shown (below 2%).

In order to quantify more accurately the degradation between both film architectures and the various enzyme concentrations used, the Voigt-based viscoelastic model was used (see Equations 1 and 2). The model takes into account the dissipation variations and thus the variations of coupled water mass during the degradation. As observed in Figures 2C and 2D, the thickness reduction of (CHT/HA)3 films was more pronounced than with (CHT/HA)3-CHT ones. In the case of (CHT/HA)3 films (Figure 2C), there was a fast degradation in the first 2h for the three highest concentrations, for which film degradation exceeded 60%. At the same time-point, film degradation was only 20-33% at lower concentrations, less than half of the higher concentrations. In the next time-points for the three highest Hase concentrations, the thickness decreases were not as pronounced as in the first 2h. For 2000 µg·mL-1, thickness reduction progressed to 97%, corresponding to a near-full degradation of the film. Among all the tested concentrations, Hase at 50 µg·mL-1 proved to be the concentration that provided thickness reduction at a steadier pace, indicates a more progressive degradation through time (in accordance to the tendency observed in the QCM-D graphics in Figure 1): after 24h, the thickness decreased 67% at this Hase concentration.

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 39

For (CHT/HA)3-CHT films, the thickness variations follow the same trend as the former architecture. However, the variations reveal that the degradation is not as pronounced as with HA-ending films. For example, for 2000 µg·mL-1 Hase the degradation only reached 68%, much lower than the 97% observed for the (CHT/HA)3 films, thus showing that the degradation of LbL films can be controlled not only by its composition but also by the type of polyelectrolyte adsorbed at the last layer. The thickness variations of films incubated in Hase-free conditions were not significant, accounting for a reduction of less than 2% in both film architectures. The quantification of the thickness reduction confirms that the lack of specificity of Hase towards the last CHT layer delays the degradation of the film when incubated with this enzyme. The “in” and “out” diffusion that is typical of LbL films with exponential growth may also help Hase access to HA cleavage points, since CHT and HA should not be structured as well-defined layers but rather as a blend of these two materials. The QCM-D results showed that the enzymatic degradation of polyelectrolyte ultrathin films depends on the type of the last layer and on the affinity between enzyme and polymer. In order to have a better understanding of films degradation profile, homogeneity and how Hase affects the film structure, AFM was used to analyze their topography and roughness. Hase at 50 µg·mL-1 was selected for the following experiments since it exhibited a more gradual degradation during 24h, while achieving about two thirds of total thickness loss (see Figure 2C). From this point on, only the (CHT/HA)3 film lacking the ending layer of CHT were studied. AFM was performed in (CHT/HA)3 films after predetermined incubation periods: 0h (used as a control), 3h, 6h, 9h and 24h in contact with Hase. Figure 3 shows the surface modification and roughness values

18


Page 19 of 39

variations in (CHT/HA)3 films during the enzymatic degradation (arithmetic average, Ra; root mean square, Rq).

A

Nondegraded

B

3 hours

1 µm

1 µm 24 hours (F)

E

20 nm

6 hours

1 µm 9 hours

D

C

Roughness (nm)

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

7

-20 nm *

Ra Rq

*

*

6

* *

*

5

* *

4

* *

3 2 1

1 µm

1 µm

0 0h

3h

6h

9h

24h

Figure 3. AFM images (3×3 µm2) of (CHT/HA)3 films (A) prior to degradation, and in contact with Hase solutions at 50 µg·mL-1 after (B) 3, (C) 6, (D) 9, and (E) 24h. (F) Roughness analysis from root mean squared roughness (Rq) and average roughness (Ra) determined from three independent samples. Data are means±SD. Significant differences were found for *p

Enzymatic Degradation of Polysaccharide-Based ... - ACS Publications

Recommend Documents