Covalent immobilization of bioactive poly(amidoamine)s onto plasma-functionalized PLGA surfaces

This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 87.2.211.120

This content was downloaded on 13/07/2014 at 09:24

Please note that terms and conditions apply.

Covalent immobilization of bioactive poly(amidoamine)s onto plasma-functionalized PLGA

surfaces

View the table of contents for this issue, or go to the journal homepage for more

2014 Mater. Res. Express 1 035001

(http://iopscience.iop.org/2053-1591/1/3/035001)

Home Search Collections Journals About Contact us My IOPscience

Covalent immobilization of bioactive poly(amidoamine)s onto plasma-functionalized PLGAsurfaces

Stefano Zanini1, Claudia Riccardi1, Antonino Natalello2,Graziella Cappelletti3, Daniele Cartelli3, Fabio Fenili4,Amedea Manfredi4 and Elisabetta Ranucci41Università degli Studi di Milano-Bicocca, Dipartimento di Fisica ‘G. Occhialini’, p.za dellaScienza, 3 I-20126 Milano, Italy2Dipartimento di Biotecnologie e Bioscienze, Università degli Studi di Milano-Bicocca, p.zadella Scienza, 2 I-20126 Milano, Italy3Dipartimento di Bioscienze, Università degli Studi di Milano, via Celoria, 26 I-20133 Milano,Italy4Dipartimento di Chimica, Università degli Studi di Milano, via Golgi, 19 I-20133 Milano, ItalyE-mail: [email protected]

Received 17 January 2014, revised 22 May 2014Accepted for publication 11 June 2014Published 30 June 2014

Materials Research Express 1 (2014) 035001

doi:10.1088/2053-1591/1/3/035001

AbstractAn approach to the surface modification of poly(lactic-co-glycolic acid) (PLGA)to render it adhesive to poly(amidoamine) (PAA) hydrogels, thus allowingfabrication of entirely biodegradable and biomimetic multilayered compositebiomaterials with the PLGA film playing the role of reinforcing material, forinstance imparting resistance to stitching, is N2/H2 plasma treatment of PLGAsurfaces aimed at introducing amine groups and covalently immobilizing PAAs.Grafting of linear PAAs, demonstrated by XPS analysis, is reported first.Coherent PAA/PLGA composite hydrogels with embedded PLGA films can beobtained likewise. They are soft, elastic and resistant to osmotic shock. Incontrast, hydrogels prepared from untreated PLGA films delaminate on swelling.Accessible hybrid PAA/PLGA materials may expand PLGA’s biomedicalapplications.

S Online supplementary data available from stacks.iop.org/MRX/1/035001/mmedia

Keywords: biodegradable, hydrogels, biomimetic, cell culturing, poly(lactic-co-glycolic acid), polyamidoamines, plasma treatments

Materials Research Express 1 (2014) 0350012053-1591/14/035001+18$33.00 © 2014 IOP Publishing Ltd

Introduction

Poly(lactic-co-glycolic acid) (PLGA) is a well-known biocompatible and biodegradablematerial used for different applications such as bioerodible implants, scaffolds for regenerativemedicine and matrices for the micro- and nanoencapsulation of therapeutics [1–3]. Despite itsotherwise remarkable properties, it is generally recognized that the hydrophobicity of PLGAsurfaces disfavors cell-adhesiveness, as indicated, among others, by the limited success ofporous PLGA scaffolds with non-modified surfaces in three-dimensional cell proliferation tests[4–6]. The hydrophobicity of PLGA prevents physical compatibility with biomimetichydrophilic polymers too, thus hampering the production of coherent multilayered compositehydrogels with embedded PLGA films [7]. Finally, due to their physico-chemical features,particularly in terms of surface charge and affinity, intravenously injected PLGA nanoparticlesare rapidly uptaken by cells of the reticuloendothelial system [8]. Many efforts have been so fardevoted to the surface modification of PLGA by a variety of chemical and physical methods,including hydrolysis [9, 10], aminolysis [10], blending with block copolymers [5, 11, 12],ozone treatment [13], ion beam irradiation [14], and plasma treatments [15–19]. The lasttechnique has been also extensively employed for introducing interfacial bonding layers for thesubsequent immobilization of cell-adhesive biomacromolecules [20–25].

Poly(amidoamine)s (PAAs) represent a family of synthetic polyelectrolytes with arecognized potential for pharmaceutical and biotechnological applications [26]. They areprepared by polyaddition of prim- or sec-amines with bis-acrylamides in water or alcohols atmoderate temperatures and usually without added catalysts. Most PAAs are water-soluble or atleast water-swellable. Since under normal conditions the polyaddition leading to PAAs isremarkably specific, a number of supplementary functions such as for instance carboxyl-,sulphonic-, hydroxyl-, additional tert-amine- and guanidine groups can be introduced as sidesubstituents, thus enabling one to prepare products whose structure and physico-chemicalproperties, such as for instance solubility and acid-base behavior in aqueous media, can betailored within wide limits [27]. In particular, PAAs can be designed to be biocompatible andbiodegradable to non-toxic products in aqueous media [28–31]. PAAs bearing disulphide bondsin the main chain have been found amenable to bio-reductive degradation inside cells and wereproposed as intracellular carriers for bioactive substances [32, 33]. An amphoteric PAA bearingcarboxyl groups as side substituents, intravenously administered in mice, exhibited ‘stealth-like’properties, and passively concentrated in solid tumors by the enhanced permeation and retentioneffect [30]. PAAs containing either thiol or phenanthroline pendants have been recentlyproposed as carriers for rhenium and/or ruthenium ions as imaging probes [34, 35].

Cross-linked PAAs form hydrogels that swell in aqueous media to a degree dependingboth on chemical structure and cross-link density, forming hydrogels. They can be prepared byusing ab initio multifunctional monomers or post-polymerizing suitably end-functionalizedoligomers, that is, short-chain polymers, for instance the acrylamide-terminated ones obtainedby employing excess bisacrylamide in the reactant mixture. Several PAA hydrogels were testedas substrates for cell culturing [36–41]. It was found that, as a rule, amphoteric, but prevailinglyanionic PAAs are biocompatible, but scarcely adhesive, whereas several cationic PAAs, besidesbeing biocompatible, are adhesive towards different cell types. Hydrogels deriving from anamphoteric but prevailingly cationic PAA, named AGMA1, were used to prepare tubularconduits for peripheral nerve regeneration [42]. Implanted in rats, these conduits showedexcellent performance as regards biocompatibility and ability of inducing peripheral nerve

2

Mater. Res. Express 1 (2014) 035001 S Zanini et al

regeneration, whereas inflammation or otherwise untoward side effects were totally absent.However, their future employment in clinics was seriously hampered by their poor mechanicalproperties. For instance, they could by no means be sutured and had to be inserted in liveanimals by gluing.

PAA grafting onto PLGA might lead to strong covalent attachment of PAA hydrogels toPLGA articles, such as fibers and sheets, thus combining the excellent biological properties ofthe former with the superior mechanical properties of the latter in entirely biodegradable newcomposite materials. To this purpose, the devised strategy was to introduce NH2 or NH groupsonto the PLGA surface, rendering it amenable to undergoing additional reaction with theterminal groups of acrylamide-terminated PAAs. Plasmas of N2/H2 mixtures, moreenvironmentally friendly and easier to handle than ammonia, were employed. According toliterature reports, N2/H2 plasmas compared with NH3 plasmas allowed introducing aminegroups onto poly(ethylene) and poly(styrene) surfaces with comparable efficiency in terms offunctionalization degree (reaching in both cases between 0.5 and 2.0 sites nm−2), but higherNH2 selectivity, that is NH2/N ratio [43, 44].

The aim of this paper is to relate on the covalent immobilization of PAAs onto amine-functionalized PLGA surfaces, in turn obtained by N2/H2 plasma treatment, and their use for thepreparation of both PLGA films surface-grafted with linear PAAs and homogeneousmultilayered cross-linked PAA/PLGA composites apparently behaving as internally-reinforcedhydrogels.

Materials and methods

Materials

2,2-Bis(acrylamido)acetic acid (BAC) [45] and N,N′-bis(acryloyl)piperazine (BP) [46] wereprepared as reported, and purity determined by titration and NMR spectroscopy. 50:50 Poly(DL-lactide-co-glycolide) (PLGA) was purchased from DURECT Corporation (Pelham, AL,United States). Taurine (TAU) was purchased from Acrös Organics (Carlo Erba, Italy). Allother reagents were purchased from Aldrich and used as-received, without further purificationsif not otherwise specified. Analytical grade HPLC solvents were purchased from Fluka and usedas- received.

Instruments

Molar mass averages, Mn and Mw, of acrylamide-end-capped and OH-terminated PAAs weredetermined by size exclusion chromatography (SEC). SEC traces were obtained with Toso-Haas TSK-gel G4000 PW and TSK-gel G3000 PW columns connected in series using a Watersmodel 515 HPLC pump equipped with a Knauer autosampler 3800, a light scattering (LS)Viscotek 270 dual detector, UV detector Waters model 486 operating at 230 nm, and arefractive index detector Waters model 2410. The mobile phase was a 0.1M Tris bufferpH 8.00 ± 0.05 with 0.2M sodium chloride. The flow rate was 1mLmin−1 and the sampleconcentration was 1% w/w. 1H spectra were run on a Brüker Advance 400 spectrometeroperating at 400.132MHz.

Polymers were purified by ultrafiltration through an Amicon® apparatus using membraneswith a nominal cut off 1000.

3

Mater. Res. Express 1 (2014) 035001 S Zanini et al

The size and ζ-potential of PLGA nanoparticles were determined by dynamic lightscattering (DLS). DLS analyses were carried out using a Malvern NanoZS instrument (MalvernInstruments, Worcestershire UK) with a laser fitted at 532 nm and fixed 173° scattering angle.The sample solutions (1mgmL−1) were filtered through 5.0 μm nylon filters.

Hydrophilic characteristics of the plasma treated PLGA films were evaluated by staticcontact angle (WCA) measurements, using a DataPhysics OCA 20 instrument.

Surface elemental analyses were determined by x-ray photoelectron spectroscopy (XPS).XPS spectra were recorded using a Perkin-Elmer PHI 5400 ESCA System apparatus, with aMg-anode Kα source and an electron take-off angle of 45°. The analyzed circular area had adiameter of 0.8mm. The pressure in the chamber was around 10−8 Pa. The spectrometer wascalibrated by using the Ag 3d5/2 peak and the resulting energetic resolution was 0.46 eV.

Amino groups surface density was determined by fluorescamine coupling [47, 48]. ThePLGA samples (2 cm× 2 cm) were dissolved in 2mL of a 9 × 10−4M solution of fluorescaminein acetone. The solution fluorescence was measured in a 1 cm path length quartz cell using theCary Eclipse Varian fluorimeter (Varian Australia Pty Ltd, Mulgrave VIC, AU) with anexcitation wavelength of 400 nm and emission at 475 nm. The amount of primary amino groupswas determined based on a calibration curve constructed by reaction of methoxypolyethyleneglycol amine (in solutions of known concentration) and fluorescamine.

Glass silane functionalization and mold preparation

Square glass plates (5 cm× 5 cm) were rendered non-adhesive by soaking for 5 h in aqua regia atroom temperature, then washing several times with water, drying and exposing tochlorotrimethylsilane vapors (20mL) overnight in a closed chamber. The plates were retrievedand washed with ethanol (2 × 50mL) and doubly distilled water (3 × 50mL). They were finallydried with soft paper.

PLGA film casting

A 10% (w/v) PLGA chloroform solution (5mL) was cast onto a 5 cm× 5 cm silanized glasssheet inside a 1mm thick silicone square frame (3.5 cm× 3.5 cm internal size) glued onto theglass sheet. Chloroform was allowed to spontaneously evaporate for 48 h at room temperature.A thin transparent PLGA film was obtained which was dried for 24 h in vacuo (0.5mbar) in athermostatic chamber at 37 °C.

Preparation of PLGA nanoparticles and PAA coating procedure

PLGA nanoparticles were prepared as reported [49]. Briefly, PLGA (20mg) was dissolved indimethylformamide (10mL). The solution was introduced into a dialysis tubing (molecular cutoff 12 000) dialyzed against distilled water (3 × 500mL) throughout 24 h then freeze-dried.PLGA nanoparticles were suspended into a 0.5% w/w aqueous solution of the cationic PAAISA1 (see below) (5mL). The pH was adjusted to 7 and the solution was kept under gentlestirring for 1 h and the excess PAA removed by dialysis (molecular cut-off 50 000) againstwater (3 × 500mL) throughout 24 h. PLGA nanoparticles were recovered by freeze drying andcharacterized by DLS and ζ-potential measurements.

4

Mater. Res. Express 1 (2014) 035001 S Zanini et al

Differently charged PAA-polyelectrolytes for the modification of PLGA were obtained byintroducing various acid and basic functions in their repeating units. The aim was to investigatethe effect of the polyelectrolyte nature on PAA/PLGA adhesion.

Synthesis of acrylamide-end capped BAC-TAU

BAC (0.68 g, 3.33mmol) and sodium hydroxide (0.13 g, 3.33mmol) were dissolved in distilledwater (1mL). Taurine (0.38 g, 3.00mmol) was added and rapidly dissolved under stirring, andthe reaction mixture kept 14 days at room temperature in the dark under nitrogen atmosphereand magnetic stirring. After this time the solution was diluted to 100mL with water and thepH adjusted to 4.5–5.0 with 37% hydrochloric acid. The product was finally purified byultrafiltration and recovered by freeze drying the retained portion. Yield: 0.66 g (62.6%).Molecular mass values: Mn = 3300Da, Mw = 4990Da, Mw Mn/ = 1.51.

Synthesis of acrylamide-end capped BAC-DTT and MBA-TAU

Acrylamide-end capped BAC-DTT was prepared in the same way as acrylamide-end cappedBAC-TAU, replacing taurine with 1,4-Di-thio-D,L-threitol (DTT) (0.43 g, 2.73mmol).Nitrogen was flushed through the reaction mixture for 30min, the pH adjusted to 6 by addingfew drops of 1M hydrochloric acid, and the solution kept 14 days at room temperature in thedark. Yield: 0.59 g (57.3%). Molecular mass values: Mn = 4890Da, Mw = 5860Da,

Mw Mn/ = 1.20.MBA-TAU was prepared as BAC-TAU, replacing BAC with methylenebisacrylamide

(MBA) (0.52 g, 3.33mmol). The reaction mixture was kept 14 days at 40 °C. Yield: 0.57 g(68.3%). Molecular mass values: Mn = 4800Da, Mw = 5470Da, Mw Mn/ = 1.14.

Synthesis of acrylamide-end capped ISA1

To a solution of BP (2.40 g, 12.36mmol) in water (3.4mL), N,N′-bis(hydroxyethyl)ethylenediamine (0.54 g, 5.15mmol) and 2-methylpiperazine (0.79 g, 5.15mmol) were added.The reaction mixture was stirred until a clear solution was obtained, and then maintained atroom temperature for 5 days. After this time, the product was isolated by diluting with water(150mL), acidified with 1M hydrochloric acid until pH 5 and ultrafiltered. The retained portionwas recovered by freeze drying. Yield: 1.52 g (40.8%). Molecular mass values: Mn = 2300Da,Mw = 3590Da, Mw Mn/ = 1.56.

Synthesis of OH-terminated BAC-TAU, MBA-TAU, BAC-DTT and ISA1

A polymer sample (200mg) (corresponding to 0.12mmol of residual double-bond for BAC-TAU, 0.14mmol for MBA-TAU, 0.11mmol for BAC-DTT and 0.073mmol for ISA1) wasdissolved in de-oxygenated doubly distilled water (1mL). The pH was adjusted to 6 by adding afew drops of 0.1M sodium hydroxide, and 2-mercaptoethanol (70 μL, 1mmol) was added. Thereaction mixture was stirred for 6 days at room temperature, then diluted with water andultrafiltered recovering by freeze drying the retained portion. Yield: 93mg (46.5%) for BAC-TAU, 62mg (31.0%) for MBA-TAU, 75mg (37.5%) for BAC-DTT and 100mg (50%) forISA1. No residual acrylamide functions were detected by 1H NMR (data not shown).

5

Mater. Res. Express 1 (2014) 035001 S Zanini et al

Plasma modification of PLGA films

The employed plasma reactor was constituted by a cylindrical glass chamber (inner diameter10 cm, length 30 cm), closed at each end with stainless steel flanges (figure 1) [48, 50]. An RFpower supplier (13.56MHz) was connected through a matching network to a copper ring placedaround the cylindrical chamber. Two other copper rings, placed apart from the RF antenna,were grounded. The PLGA film was positioned as displayed in figure 1. Prior to every plasmatreatment, the reactor was evacuated up to 0.1 Pa by a rotary pump. After this step, nitrogen andhydrogen were introduced in the chamber. The nitrogen flow was measured directly through aflow meter (EL-Flow series F-201C by Bronkhorst) while hydrogen was introduced through aneedle valve. In order to maximize the surface density of inserted amino groups, differenttreatment conditions were tested, varying the principal operating parameters (N2 flow ratebetween 2 sccm and 6 sccm, N2:H2 partial pressure ratio, power input between 15W and 25W).After the plasma treatment (exposure time varied between 1 and 4min), PLGA samples wererecovered for characterization.

Study of the behavior of plasma-treated PLGA in water solutions

A PLGA film was plasma treated using the operating parameters (power input, treatment time,N2/H2 partial pressure ratio) which maximized the surface insertion of amino groups. Thetreated film was then put in a NaHCO3 solution (pH∼ 8.5) and XPS analyses were performedafter different immersion times (ranging from 1min to several hours). The N/C ratio, calculatedfrom XPS results, was taken as an indicator of the stability of the surface functionalization inthe aqueous solution.

Chemical immobilization of PAAs onto plasma treated PLGA films

It is well documented that when a polymeric material is functionalized with NH2 groupsthrough plasma grafting, it suffers from a time-dependent decay of [N] at the topmost surface

Figure 1. Schematic of the plasma reactor used for plasma functionalization of PLGAfilms.

6

Mater. Res. Express 1 (2014) 035001 S Zanini et al

after storage in ambient air after roughly one week, on account of ‘reptational’ motion of near-surface macromolecules [51]. Moreover, the concentration of primary amino groups on thetopmost surface can decay also by reaction with atmospheric oxygen to form amides, asdisplayed by Wertheimer et al for NH2-containing plasma coatings [52, 53].

In order to overcome the ageing phenomenon and to maximize the NH2 groups availablefor the PAA immobilization, the plasma treated PLGA film was recovered from the plasmareactor and immediately dipped in a freshly prepared 10% (w/w) aqueous solution of theacrylamide-end-capped PAA and the pH adjusted to 8.5 by means of NaHCO3. After 24 hreaction time, the film was retrieved, extensively washed with doubly bidistilled water(4 × 100mL), and finally dried at room temperature. Parallel reactions with OH-terminatedPAAs were performed following the same procedure.

Cell culture

For cell culture experiments, PLGA substrates were deposited on silicon-treated 24-well tissueculture plates, and sterilized overnight by UV treatment. The day after, human lung carcinomacell line A549 (CCl-185; American Type Culture Collection, Rockville, MD, USA) wereseeded at the density of 25 000 cell/well, and were grown in minimal essential medium withEarle’s (E-MEM), supplemented with 10% fetal bovine serum (Hyclone Europe, Oud-Beijerland, Holland), 2mM l-glutamine, 100UmL−1 penicillin, and non-essential amino acids.Cells were maintained at 37 °C in a humidified atmosphere at 5% CO2, to permit cell adhesion,3 h later, cultures were washed to remove unattached cells and then analyzed byimmunocytochemistry. A549 cells were fixed and permeabilized for 10min with methanol at−20C, washed with PBS, and blocked in PBS + 1% bovine serum albumin (BSA) for 15min atroom temperature. To localize tubulin, the cells were incubated with monoclonal anti-α-tubulinantibody (clone B-5-1-2, Sigma–Aldrich, St. Louis, MO) 1:500 in PBS for 1 h at 37 °C. Weused goat anti-mouse Alexa Fluor™ 594 (Molecular Probes, Eugene, OR) 1:1000 in PBS + 5%BSA for 45min at 37 °C as secondary antibodies. Nuclei staining was performed by incubationwith DAPI (0.25 μgml−1 in PBS) for 15min at room temperature The coverslips were mountedin Mowiol® (Calbiochem)–DABCO (Sigma–Aldrich) and examined with the Axiovert 200Mmicroscope (Carl Zeiss, Oberkochen, Germany), at 63x magnification. To quantify adherentcells, after washing we added a fresh culture medium containing 10% of the Cell ProliferationFluorescence Assay Reagent (Cell Proliferation Assay Kit, Fluorimetric-Blue, abcam); after afurther 3 h of incubation, fluorescence intensity was estimated by a plate reader (Infinite®F200PRO, TECAN, Männedorf, Switzerland). The statistical significance of treatment wasassessed by one-way ANOVA, Tukey HSD post-hoc test, and analyses were performed usingSTATISTICA (StatSoft Inc., Tulsa, OK). All analyses were carried out in triplicate.

Synthesis of ISA1-PLGA hydrogels

ISA1 acrylamide-end-capped oligomer (1 g, 0.37mmol acrylamide functions) was dissolvedunder nitrogen atmosphere in doubly distilled water (1.5mL) and 0.5mL of an aqueous solutionof 4,4′-azobis(4-cyanovaleric) acid (250mg in 5mL doubly distilled water) added. The mixturewas stirred for 1min, then retrieved with a syringe and injected in a mold consisting of6 cm× 6 cm glass previously treated with trimethylchlorosilane, containing a plasma treatedPLGA film of 5 cm× 5 cm between 2 not-adhesive polyethylene membrane of 0.5mm each.The reactive mixture was then exposed to UV radiation for 2 h. The hydrogel samples were

7

Mater. Res. Express 1 (2014) 035001 S Zanini et al

finally recovered and soaked in water (2 × 200mL) and in 0.1mM PBS pH 7.0 (1 × 200mL),each time for 4 h. The same procedure was adopted with the untreated PLGA films.

Rheological measurements

Viscoelastic measurements were performed using a strain-controlled rheometer (RheometricScientific ARES). Samples were prepared in the form of disks with a 28mm diameter and 1mmthickness in the swollen state. Samples were tested using a parallel plate geometry with a 25mmdiameter. The samples were placed between the rheometer plates, and a slight compressiveforce of about 50 g was applied. The frequency sweep test was conducted with a strain of 0.2%from 0.05 to 100Hz at 25 °C.

Swelling tests

Native hydrogels were cut into 10 × 10 × 0.3mm3 parallelepiped, freeze-dried (mean weight212 ± 15mg) and placed into a Falcon tube containing 50mL of medium, either water, orphosphate buffer pH 7.4 or ethanol at room temperature for 24 h. The swelling percentage wascalculated using the following formula:

= ×Swelling (%) 100W

Wwet

dry

where Wwet is the weight of the swollen hydrogel and Wdry is the weight of the dryhydrogel.

Results and discussion

Synthesis of polymers

The research objective was to establish a synthetic procedure for grafting PAAs onto PLGA,thus obtaining PLGA/PAA hybrid materials combining good mechanical properties, withbiocompatibility, degradability and biomimetic properties, to be used as scaffolds for in vitroand in vivo tissue engineering studies. The selected procedure was to introduce prim- or sec-amine groups onto the PLGA surface by N2/H2 plasma treatment, followed by additionreaction with acrylamide-end-capped PAAs (figure 2 and scheme 1).

For the grafting of linear PAAs, the excess PAA was eliminated with water. PAAhydrogel grafting was attained by the same technique, but in this case the excess PAA wascross-linked with UV irradiation in the presence of radical initiators.

Cationic polymers are known to give strong polyelectrolyte interactions with thenegatively charged surface of PLGA, which stems from the partial hydrolysis of the backboneester bonds. This feature has prompted many authors to adsorb onto PLGA stable cationicpolymer layers, as in the case of chitosan and laminine coatings onto negatively chargedPLGA scaffolds to enable cells attachment [24], or in the case of poly(ethyleneimine), used tobind DNA strains onto PLGA nanoparticles [54]. Many PAAs are polycations and, therefore,are liable to be surface-adsorbed onto PLGA, giving false positive results. In fact, they gavestable coating onto PLGA nanoparticles by simply incubating them in dilute aqueous PAAsolutions. In typical experiments, PLGA nanoparticles with average size 50–200 nm wereobtained by nanoprecipitation from dilute dimethylformamide solutions into water by dialysismethods (see the experimental section). These nanoparticles exhibited negatively chargedsurfaces (ζ − potential = −20mV), which turned into steadily positive (ζ − potential = +20mV)

8

Mater. Res. Express 1 (2014) 035001 S Zanini et al

Figure 2. Schematic representation of PLGA and its modified derivatives. (a) nativePLGA, (b) plasma treated PLGA, (c) PLGA grafted with PAAs.

Scheme 1. Synthesis of acrylamide and hydroxyl end-capped poly(amidoamine)s.

9

Mater. Res. Express 1 (2014) 035001 S Zanini et al

after only very short contact times, in the order of a few minutes, with dilute solutions ofdifferent cationic PAAs, including the one known as ISA1, obtained from bisacryloylpiper-azine, 2-methylpiperazine and N,N′-dihydroxyethyl-ethylenediamine (scheme 1). Thispositive charge persisted even after extensive washing with distilled water and the PAAcoating disappeared only with the occurrence of PLGA degradation.

In order to unambiguously demonstrate the covalent binding of PAAs onto PLGAsurface, thus ruling out the dominant effect of ionic interactions of the PAA chains withPLGA surfaces, two anionic acrylamide end-capped PAAs, namely BAC-TAU and MBA-TAU, were first chosen as models for studying the grafting reaction (scheme 1). Thepresence of one taurine residue per repeating unit imparted a net anionic character to bothBAC-TAU and MBA-TAU, due to the strong acid dissociation of the sulfonic groups and,in addition, the presence of one sulphur atom per repeating unit provided a tracing probe forthe subsequent XPS elemental analysis of the PAA modified PLGA surfaces. Two differentbisacrylamides, namely BAC and MBA, were used as monomers, characterized by differentreactivity in the addition reaction with amines, combined with different hydrophilic/hydrophobic balance and net average negative charge of the resulting polymers. Inparticular, the repeating unit of BAC-TAU contained three ionizable groups, two strongacids (pKa1 < 1 and pKa2 = 2.1) and a medium-strength base (pKa3 = 7.1) [27], whereas thatof MBA-TAU contained only two ionizable groups, a strong acid and a medium-strengthbase whose pKas were estimated, in comparison with those of related polymers, as pKa1 < 1and pKa2 = 7.4 [27]. Therefore, both BAC-TAU and MBA-TAU were prevailingly anionicat pH 7.4 with an average excess negative charge of 1.5 and 0.5 per unit, respectively. Inaddition, in order to increase the sulphur content in the grafted polymer, thus improving thedetection of the grafted layer, a third model polymer, the poly(thioetheramide) BAC-DTT(scheme 1), was synthesized by polymerization of 1,4-di-thio-D,L-threitol (DTT).

The reaction conditions used for grafting the anionic BAC-TAU, MBA-TAU and BAC-DTT onto plasma-modified PLGA surfaces were finally adopted to covalently immobilize thecationic ISA1. Besides being biocompatible, ISA1 has proved to be active in intracellulartrafficking [26], and cell-adhesive towards different cell lines [55].

End-capped BAC-TAU, MBA-TAU, BAC-DTT and ISA1 were prepared according toa conventional procedure, consisting of the polyaddition of amine monomers or DTT with a10% molar excess of bisacrylamide, either BAC or MBA or BP (scheme 1). Thepolymerization reactions took place in aqueous solution, and in order to rule out thepresence in the final product of very short reactive oligomers, the raw mixtures werepurified by ultrafiltration through membranes with nominal cut-off 1000. The presence ofacrylamide residues on chain terminals was ascertained by means of NMR spectroscopy.The final products were oligomers with relatively low molecular masses (ranging from5000 Da to 6000 Da), consistent with the 10% stoichiometric unbalance between the amineor thiol/acrylamide functions in the starting reactive mixtures. In order to definitely excludeany non-covalent interactions between the anionic oligomers and PLGA surface, theunreactive OH-terminated counterparts of BAC-TAU, MBA-TAU, BAC-DTT and ISA1were obtained by reaction with excess 2-mercaptoethanol and incubated with amine-functionalized PLGA films under the same conditions adopted in previous graftingexperiments.

10

Mater. Res. Express 1 (2014) 035001 S Zanini et al

Plasma modification of PLGA films

Plasma treatment of polymeric materials with N2/H2 plasmas generally leads to the surfaceincorporation of different N-containing functions [45, 46]. In this work, plasma treatment ofPLGA films was preferentially performed using a 1:2 N2/H2 ratio, reported as one of the mostefficient for introducing primary amine groups [44, 48], and varying the other relevantoperating parameters (table 1). The extent of functionalization under the adopted conditions wasdetermined by XPS analysis, whose survey spectra are reported in figure 3(a) both for theuntreated and plasma treated PLGA films. The presence of the N1s peak was observed only inthe spectrum of the treated sample, thus confirming the occurrence of grafting of N-containingfunctions onto PLGA after plasma treatment.

The chemical surface composition of four representative samples is reported in table 1.

(a) (b)

Figure 3. (a) XPS survey spectra of untreated and plasma treated PLGA films; (b) highresolution XPS spectrum of plasma treated PLGA film showing the N1s peak.

Table 1. Operating parameters, surface composition from XPS analysis and contactangle measurements of differently prepared PLGA samples.

Sample

N2

flow(sccm)

PN2

(Pa)Ptot(Pa)

PowerInput(W)

Treatmenttime (min) C O N O/C N/C

Contactangle (°)

PLGA — — — — — 59.3 39.0 — 0.66 — 74PLGA-NH2-1

2 6 19 15 4 62.8 34.0 1.9 0.54 0.030 35

PLGA-NH2-2

7 13 40 15 4 70.0 24.2 2.5 0.35 0.036 56

PLGA-NH2-3

7 13 40 25 4 69.4 19.0 3.1 0.27 0.45 24

PLGA-NH2-4

7 13 40 25 1 66.7 28.6 1.8 0.43 0.027 n.m.

11

Mater. Res. Express 1 (2014) 035001 S Zanini et al

These results allowed drawing different conclusions. The N/C ratio of the treated samplesranged between 0.027 and 0.045, depending on the adopted operating parameters, and from aquantitative point of view these results are similar to those reported for NH3 plasma[17, 20, 56]. As expected, the N/C ratio increased by increasing the N2 flow, the power inputand the treatment time. The O/C ratio markedly decreased after plasma exposure following theseverity of treatment and proceeded in parallel with N/C increase. This was likely due to apartial fragmentation and decarboxylation of PLGA caused by plasma bombardment. Thewettability of the treated samples, as determined by water contact angle measurements,confirmed the increased hydrophilicity of surfaces following the introduction of nitrogen-containing functions (table 2).

It is well known that one of the most important problems concerning plasmafunctionalization of polymers is the excess of energy delivered, which negatively affect theselectivity of the process [57–59]. For example, the selectivity towards primary amine groupsafter ammonia plasma exposure was found to be <10% of all inserted nitrogen-functions.Therefore, the N/C ratio obtained from XPS analysis (which is related to the N-insertionefficiency of the process) may not be regarded as evidence of the exclusive presence of thedesired primary and secondary amine functions [58]. The N1s high resolution peak of thetreated PLGA (figure 3(b)) is centered at about 401 eV, indicating that at least the 80% of thenitrogen is present as quaternary nitrogen. Only a shoulder at low B.E. (399 eV) can beattributed to primary amine.

A general procedure to quantitatively evaluate at least the primary amine content ontoplasma modified polymer surfaces makes use of the fluorescence dye fluorescamine, a non-fluorescent compound that selectively reacts with primary amine groups yielding a fluorescentproduct [47, 48]. In this work, fluorescamine test was carried out in acetone solution, being bothPLGA and fluorescamine soluble in this solvent. The fluorescence spectra of an untreated andof a plasma treated PLGA film are shown in figure 4(a). It may be noticed that the curve of theplasma treated sample exhibited one single emission peak at 478 nm and was very similar tothose reported for the reaction products of fluorescamine with amine surface modified syntheticpolymers [47].

This confirmed the presence of fluorescence emitting compounds produced by the reactionof fluorescamine with the NH2 groups present onto PLGA. The curve of the untreated sampleexhibited two clearly less intense peaks, placed at 448 and 472 nm. The fluorescence emission

Table 2. XPS analyses of PLGA films after grafting with acrylamide end-capped andOH-terminated poly(amidoamine)s.

Sample O N C S

PLGA-NH2-3 19.0 3.1 69.4 n.d.PLGA-NH2-3 washed 22.7 1.5 72.0 n.d.PLGA-NH2-3 +BAC-TAU 36.3 0.4 62.2 n.d.PLGA-NH2-3 +BAC-TAU-OH 28.0 1.1 65.1 n.d.PLGA-NH2-3 +MBA-TAU 26.0 3.4 66.2 0.5PLGA-NH2-3 +MBA-TAU-OH 36.8 0.4 62.5 n.d.PLGA-NH2-3 +BAC-DTT 34.1 0.4 65.2 n.d.PLGA-NH2-3 +BAC-DTT-OH 26.3 1.4 68.4 n.d.PLGA-NH2-3 + ISA1 22.0 3.4 73.4 n.d.PLGA-NH2-3 + ISA1-OH 21.9 4.0 72.1 n.d.

12

Mater. Res. Express 1 (2014) 035001 S Zanini et al

of the fluorescamine solution being negligible, they were attributed to scattering. Thisphenomenon, which became more pronounced with increasing PLGA concentration (data notshown), is not uncommon for colloidal solutions, and may be accentuated by the poor solubilityof the very high molecular weight PLGA fractions. The strong emission observed at theexcitation wavelength (400 nm, data not shown) confirmed this hypothesis. The fluorescenceintensity of the treated sample, measured at 475 nm, allowed determining the surface density ofprimary amine groups by difference with that of the untreated one. The densities obtained withdifferent N2/H2 ratios (1:1, 1:2 and 1:3) are shown in figure 4(b). In agreement with literaturereports [44, 47], 1:2 and 1:3 N2/H2 mixtures turned to be the most efficient for introducingprimary amino groups, leading to a density of about 1.3 groups nm−2. Previous literature worksproved that this is likely due to the high formation of NH radicals in the plasma phase, whichoccurs at %H2 above the 60% [44].

Behavior of functionalized PLGA films in basic aqueous solutions

A PLGA film was treated with the best plasma parameters (parameters of PLGA-NH2-3 intable 1, 1:2 N2/H2 ratio), in order to maximize the initial surface density of primary aminegroups. The chemical liability of N2/H2 plasma treated PLGA surfaces when immersed in basicaqueous solutions mimicking the grafting conditions was ascertained by monitoring with XPSthe variation of the N/C ratio with time (figure 5).

The N/C% ratio reduced from a value of 0.049 to a ‘stable’ value of 0.023, where ‘stable’refers to the level of retained nitrogen functions following the removal of low-molecular-weightfragments produced by treatment-induced PLGA degradation [51]. Dissolution of the producedfragments was likely to be favored by the hydrophilicity following amine-functionalization oftop layers (see wettability data, table 1).

PAA immobilization onto functionalized PLGA films

Immobilization of PAAs was performed on PLGA films treated with the best plasma parameters(parameters of PLGA-NH2-3 in table 1). Acrylamide end-capped BAC-TAU, MBA-TAU,BAC-DTT and ISA1 were grafted onto amine-functionalized PLGA surfaces by immersingPLGA films in 10% (w/w) aqueous solutions of the reactive polymers, followed by extensive

Figure 4. Fluorescence emission spectra of untreated and plasma treated PLGA filmsafter fluorescamine labelling (a) and fluorescence intensity at 475 nm of samples treatedwith different N2/H2 partial pressure ratios (b).

13

Mater. Res. Express 1 (2014) 035001 S Zanini et al

washing with distilled water to remove the excess reagents. All modified films were analyzed byXPS analysis, in order to assess the presence of nitrogen and sulfur atoms arising from thegrafted polymers. Results are summarized in table 2. For comparison, the chemical surfacecompositions of the PLGA-NH2-3 sample before and after immersion in pH 8.5 NaHCO3

aqueous solution are also reported. The presence of sulfur atoms was observed only on thesurface of the MBA-TAU grafted films.

The lack of reaction of BAC-based end-capped oligomers in the addition reaction with theamine functions of modified PLGA surface was ascribed to the electrostatic repulsion betweenthe carboxylate residue present on the oligomer terminals and the negatively charged PLGAsurface. No reaction occurred in the case of sulfur-containing OH-terminated PAAs, whichproved unreactive towards addition reactions. In particular, the different behavior ofacrylamide- and OH-terminated MBA-TAU allowed ruling out any kinds of non-covalentinteractions, thus confirming the occurrence of the chemical immobilization of acrylamide-terminated MBA-TAU.

Differently from the anionic PAAs, the cationic PAA ISA1 is physically adsorbed ontountreated PLGA (see table 2). Therefore, it can be concluded that the amount of acrylamide-terminated ISA1 grafted onto the plasma modified PLGA (evaluated by the atomic percentageof nitrogen) is comparable with that of OH-terminated ISA1 which is adsorbed onto theunmodified films.

The biocompatibility of the PLGA surfaces was assayed using the epithelial-like A549 cellline. The adhesion on PLGA surfaces was investigated on native substrate or following plasmatreatment and PAAs grafting. Cell adhesion was assessed by means of microscopy followingthe staining for tubulin and nucleus. After 3 h of adhesion, cells were attached and seemed quiteenlarged on both native and modified PLGA surfaces (online supplementary figure S1(a)available at stacks.iop.org/MRX/1/035001/mmedia) with the exception of MBA-TAU, the mostanionic polymer in the series, which probably induces repulsive interactions. Furthermore, cellsseeded on PLGA treated with plasma showed a more pronounced spreading with respect tocontrol culture. The evaluation of adherent cells on the differently treated PLGA surfaces withrespect to controls allowed definitely demonstrating that plasma treatment significantly

Figure 5. N/C ratio of PLGA-NH2-3 film after incubation in pH 8.5 NaHCO3 aqueoussolution.

14

Mater. Res. Express 1 (2014) 035001 S Zanini et al

ameliorated cell adhesion within the first 3 h (supplementary figure S1(b)), whereas nosignificant differences were observed in cell adhesion after incubating for 24 h (data not shown).

Synthesis of PAA/PLGA composite hydrogels

Composite hydrogels made of 0.5mm ISA1 layers intercalated with a thin PLGA film wereprepared by dipping a 0.25mm thick amine-modified PLGA film in a acrylamide-end-cappedISA1 solution, inside a 1mm thick 5 cm× 5 cm mold and then curing under UV irradiation inthe presence of a radical initiator. Under these conditions, the prim- and sec-amine groups onPLGA surface could participate in the addition reaction with terminal acrylamide double bonds.The ISA1/PLGA composite hydrogel was purified, as normally done, by extensive washingwith water. The resulting product was a transparent, soft, elastic and macroscopicallyhomogeneous material, which exhibited a high extent of swelling in aqueous media or polarsolvents (figure 6).

In particular, maximum swelling was observed in water and minimum in ethanol, inagreement with the high hydrophilicity of ISA1 hydrogels. Interestingly, ISA1/PLGA hydrogelshowed a slightly lower swelling in aqueous media than the unreinforced ISA1 hydrogel,consistent with the stiffening effect of the PLGA film, and a higher swelling extent in ethanol,in agreement with the expected higher lipophilicty of the PLGA reinforced hydrogel.

No sign of detachment among the ISA1 and PLGA layers was observed upon repeatedswelling (figure 7(a)) and de-swelling processes. By contrast, ISA1/PLGA hydrogels preparedfollowing the same procedure but using unmodified PLGA films appeared somewhat opaqueand evidently delaminated after prolonged immersion in water (figure 7(b)). A differentdegradation behavior was observed too. ISA1/PLGA hydrogels from treated PLGA followed adegradation pattern in phosphate buffer pH 7.4 very similar to that of normal PAA hydrogels,with a progressive degradation mediated by increased swelling and no impairment of the opticaltransparency, at least within the observation times adopted (in the range of a few days). Bycontrast, ISA1/PLGA hydrogels from untreated PLGA became progressively opaque andexhibited an increasingly irregular delamination within the same time span (figure 7(c)).

In order to ascertain the cohesiveness of the PLGA/PAA layers in the hydrogel composite,both ISA1 and ISA1-g-PLGA hydrogels were rheologically characterized by strain sweep

Figure 6. Swelling extent of ISA1 and ISA1/PLGA hydrogels.

15

Mater. Res. Express 1 (2014) 035001 S Zanini et al

analysis in which 1.25mm thick samples were deformed at 0.2% shear strain from 0.5 to100Hz frequency, while a compressive force of 0.5N was applied. A solid-like behavior wasobserved (storage modulus, G′, ≫ loss modulus, G″), indicating the formation of well-developed cross-linked and mechanically robust network. The behavior of the ISA1/PLGAcomposite hydrogel did not significantly differ from that of ISA1 one, with storage modulus,G′, values at 100 rad sec−1 of 34.6 and 28.3KPa for the ISA1/PLGA- and ISA1 hydrogel,respectively. This result was ascribed to the thickness of the hydrogel layers (≅500 μm against≅250 μm for the PLGA film), which apparently ruled the composite hydrogel response to shearstress. At the same time, the embedded PLGA film remarkably reinforced the ISA1/PLGAcomposite hydrogel, inducing an evident resistance to stitching. Up to this point, the main aimof the present study was apparently fulfilled.

Conclusions

The objective of this research work was to study the immobilization of poly(amidoamine)s ontoamine-functionalized PLGA surfaces obtained by N2/H2 plasma treatment with the ultimate aimof obtaining composite PLGA/PAA hydrogels materials hopefully combining the bestproperties of both components. The effect of plasma modification, assessed by XPSspectroscopy, revealed a sufficient level of amine functionalization, as evidenced by thepresence of N1s peaks. This determined an increase of surface hydrophilicity, as demonstratedby water contact angle measurements. The density of prim-amine groups onto plasma treatedfilms was determined by means of fluorescence dye assay. Not unexpectedly, nitrogen contentreduced to a plateau value after washing with the aqueous solutions, likely due to the removal oflow-molecular-weight polymer fragments produced by plasma bombardment. Nonetheless, theextent of surface modification was sufficient to covalently immobilize both cationic and anionicpoly(amidoamine)s.

Cell culturing experiments carried out with A549 cell line showed that poly(amidoamine)grafting did not impair PLGA biocompatibility and did not prejudge normal epithelial–like cellmorphology. In addition, all poly(amidoamine)-grafted surfaces exhibited cell adhesivenesscomparable to that of the control.

Amine-plasma modified PLGA films proved suitable to reinforce PAA hydrogels, andproducing coherent composite hydrogels, which did not delaminate upon swelling, showed asolid-like elastic behavior and were resistant to stitching.

Figure 7. Swollen ISA1/PLGA hydrogel samples obtained from a N2/H2 plasmatreated 0.3 mm thick PLGA film after 3 days immersion (a), and from an untreatedPLGA film after 10min (b) and 3 days (c).

16

Mater. Res. Express 1 (2014) 035001 S Zanini et al

On the whole, this work demonstrates the feasibility of the devised procedure to covalentlyimmobilize bioactive poly(amidoamine)s of different architectures onto plasma-activated PLGAsurfaces. This may allow fine tuning of the physico-chemical and biological properties of PLGAby combining the good biodegradability and biocompatibility of PLGA with the chemicalversatility and biomimetic properties of poly(amidoamine)s, thus widening the scope of PLGAas biomaterial.

Acknowledgment

We gratefully acknowledge FONDAZIONE CARIPLO, under project Scientific andTechnological Research for Advanced Materials programme, call 2008 for financial support.

References

[1] Ratner B D, Hoffman A S, Schoen F J and Lemons J E 2004 Biomaterials Science, An Introduction toMaterials in Medicine 2nd edn (San Diego, CA: Academic)

[2] Kulkarni R K, Mooro E G, Hegyeli A F and Leonard F 1971 J. Biomed. Mater. Res. 5 169[3] Shoichet M S 2010 Macromolecules 43 581–91[4] Murphy W L, Kohn D H and Mooney D J 2000 J. Biomed. Mater. Res. 50 50[5] Karp J M, Shoichet M S and Davies J E 2003 J. Biomed. Mater. Res. A 64 388[6] Bouhadir K H and Mooney D J 2001 J. Drug Target 9 397[7] Lawrence B J, Maase E L, Lin H K and Madihally S V 2009 J. Biomed. Mater. Res A 88 634[8] Illum L, Jacobsen L O, Muller R H, Mak E and Davis S S 1987 Biomaterials 8 113[9] Parka G E, Pattison M A, Park K and Webster T J 2005 Biomaterials 26 3075

[10] Croll I T, O’Connor A J, Stevens G W and Cooper-White J J 2004 Biomacromolecules 5 463[11] Chenga J, Teplya B A, Sherifia I, Sunga J, Luthera G, Gua F X, Levy-Nissenbauma E,

Radovic-Morenob A F, Langera R and Farokhzad O C 2007 Biomaterials 28 869[12] Makadia H K and Siegel S J 2011 Polymers 3 1377[13] Zodrow K R, Schiffman J D and Elimelech M 2012 Langmuir 28 13993[14] Tanaka T, Tsuchiya K, Yajima H I, Suzuki Y and Fukutome A 2011 Nucl. Instrum. Meth. B 269 2130[15] Yang J, Shi G, Bei J, Wang S, Cao Y, Shang Q, Yang G and Wang W 2002 J. Biomed. Mater. Res. 62 438[16] Djordjevic I, Britcher L G and Kumar S 2008 Appl. Surf. Sci. 254 1929[17] Gugala Z and Gogolewski S 2006 J. Biomed. Mater. Res. 76A 288[18] Wang Y Q, Qu X, Lu J, Zhu C F, Wan L J, Yang J L, Bei J Z and Wang S G 2004 Biomaterials 25 4777[19] Park H, Lee K Y, Lee S J, Park K E and Park W H 2007 Macromol. Symp. 249-250 103[20] Shen H, Hu X, Yang F, Bei J and Wang S 2009 Biomaterials 30 3150[21] Chung H J, Kim H K, Yoon J J and Park T G 2006 Pharm. Res. 23 1835[22] Shen H, Hu X, Yang F, Bei J and Wang S 2007 Biomaterials 28 4219[23] Shen H, Hu X and Wang S 2008 Biomaterials 29 2388[24] Huang Y C, Huang C C, Huang Y Y and Chen K S 2007 J. Biomed. Mater. Res. A 82 842[25] Yang J, Wan Y, Yang J, Bei J and Wang S 2003 J. Biomed. Mater. Res. 67A 1139[26] Ferruti P 2013 J. Polym. Sci. Part A: Polym. Chem. 51 2319[27] Ranucci E, Ferruti P, Lattanzio E, Manfredi A, Rossi M, Mussini P R, Chiellini F and Bartoli C 2009

J. Polym. Sci. Part A: Polym. Chem. 47 6977[28] Ferruti P, Franchini J, Bencini M, Ranucci E, Zara G P, Serpe L, Primo L and Cavalli R 2007

Biomacromolecules 8 1498

17

Mater. Res. Express 1 (2014) 035001 S Zanini et al

[29] Cavalli R, Bisazza A, Sessa R, Primo L, Fenili F, Manfredi A, Ranucci E and Ferruti P 2010Biomacromolecules 11 2667

[30] Richardson S C W, Ferruti P and Duncan R 1999 J. Drug Target. 6 165[31] Ranucci E, Spagnoli G, Ferruti P, Sgouras D and Duncan R 1991 J. Biomater. Sci., Polym. Ed. 2 303[32] Emilitri E, Ranucci E and Ferruti P 2005 J. Polymer Sci. Part A: Polym. Chem. 43 1404[33] Cohen S, Coue G, Beno D, Korenstein R and Engbersen J F 2012 Biomaterials 33 614[34] Donghi D, Maggioni D, D’Alfonso G, Amigoni F, Ranucci E, Ferruti P, Manfredi A, Fenili F, Bisazza A and

Cavalli R 2009 Biomacromolecules 10 3273[35] Maggioni D et al 2012 Inorg. Chem. 51 12776[36] Ferruti P, Bianchi S, Ranucci E, Chiellini F and Caruso V 2005 Macromol. Biosci. 5 613[37] Ferruti P, Bianchi S, Ranucci E, Chiellini F and Piras A M 2005 Biomacromolecules 6 2229[38] Jacchetti E, Emilitri E, Rodighiero S, Indrieri M, Gianfelice A, Lenardi C, Podestà A, Ranucci E,

Ferruti P and Milani P 2008 J. Nanobiotechnol. 14 6[39] Emilitri E, Guizzardi F, Lenardi C, Suardi M, Ranucci E and Ferruti P 2008 Macromol. Symp. 266 41[40] Dos Reis G et al 2010 Macromol. Biosci. 10 842[41] Mauro N, Manfredi A, Ranucci E, Procacci P, Laus M, Antonioli D, Mantovani C, Magnaghi V and Ferruti P

2013 Macromol. Biosci. 13 332[42] Magnaghi V, Conte P, Pivato G, Cortese P, Cavalli E, Pajardi G, Ranucci E, Fenili F, Manfredi A and

Ferruti P 2011 J. Biomed. Mater. Res., Part A 98 19[43] Meyer-Plath A A, Finke B, Schröder K and Ohl A 2003 Surf. Coat. Technol. 174-175 877[44] Wang M J, Chang Y I and Poncil-Epaillard F 2003 Langmuir 19 8325[45] Ferruti P, Ranucci E, Trotta F, Gianasi E, Evagorou G E, Wasil M, Wilson G and Duncan R 1999 Macromol.

Chem. Phys. 200 1644[46] Ferruti P 1985 Macromol. Synth. 9 25[47] Ivanov V B, Behnisch J, Hollander A, Mehdorn F and Zimmermann H 1996 Surf. Interface Anal. 24 257[48] Russo L, Zanini S, Riccardi C, Nicotra F and Cipolla L F 2011 Mater. Today 14 164[49] Jeon H J, Jeon Y I, Jang M K, Park Y H and Nah J W 2000 Int. J. Pharm. 207 99[50] Zanini S, Ziano R and Riccardi C 2009 Plasma Chem. Plasma Proc. 29 535[51] Truica-Marasescu F, Jedrzejowski P and Wertheimer M R 2004 Plasma Process. Polym. 1 153[52] Truica-Marasescu F and Wertheimer M R 2008 Plasma Process. Polym. 5 44[53] Ruiz J-C, St-Georges-Robillard A, Theresy C, Lerouge S and Wertheimer M R 2010 Plasma Process. Polym.

7 737[54] Bivas-Benita M, Romeijn S, Junginger H E and Borchard G 2004 Eur. J. Pharm. Biopharm. 58 1[55] Fenili F, Gienfelice A, Dos Reis G, Lenardi C, Ranucci E, Ferruti P and Milani P 2008 15th CIRMIB

Biomaterials School—Strategies and Materials for Regenerative Medicine: Focus on Bone and Cartilage(Ischia, Italy)

[56] Hu Y, Winn S R, Krajbich I and Hollinger J O 2003 J. Biomed. Mater. Res. 64A 583[57] Friedrich J F, Mix R, Schulze R D, Meyer-Plath A, Joshi R and Wettmarshausen S 2008 Plasma Process.

Polym. 5 407[58] Siow K S, Britcher L, Kumar S and Griesser H J 2006 Plasma Processes Polym. 3 392[59] Wettmarshausen S, Min H, Unger W, Jäger C, Hidde G and Friedrich J 2011 Plasma Chem. Plasma Process.

31 551

18

Mater. Res. Express 1 (2014) 035001 S Zanini et al