WIXLIB

Poly(amido amine) Based Multilayered Thin FilmsUV characterization of multilayers was performed in thedry state using a UV-2401 PC (Shimadzu, ‘s-Hertogenbosch,The Netherlands) UV spectrophotometer. Each film fabricated on UV-transparent 7.5 37 1 mm quartz glass (Ted Pella, Redding, USA) was measured in three different arbitrarypositions. Absorbance scan was carried out in the 200–700 nmwavelength range. All data points were then corrected forbaseline offset by subtracting the absorbance value at400 nm from each data point. Relative absorbance valueswere obtained by normalizing each data point with the respective value at time 0.Poly-D-lysine-coated 96 well plates (PDL-TCPS) for multilayer build-up for cell culture and transfection experimentswere purchased from Greiner (Alphen aan den Rijn,The Netherlands).COS-7 cells (European Collection of Animal Cell Cultures(ECACC) Catalogue No. 87021302) were grown in DMEMcontaining 4.5 g/L glucose and GlutaMAXTM (Invitrogen,Breda, The Netherlands) supplemented with 2% (v/v)PennStrepp (Lonza, Breda, The Netherlands) and 10% (v/v)fetal bovine serum (Lonza, Breda, The Netherlands).Fluorescence microscopy was performed at 4X, 10X, 20X,and/or 40X objectives using EVOS digital inverted microscope (EMS, Wageningen, The Netherlands).Confocal microscopy was performed on an LSM 510 (CarlZeiss, Sliedrecht, The Netherlands) using the Zen 2009software.Fluorescence-activated cell sorting (FACS) was carried outin a Becton-Dickinson FACSCalibur (Breda,The Netherlands).Multilayered Thin Film Construction and Build-UpProfilesFresh BA-PAA solutions were prepared shortly before thestart of multilayer build-up from the solid materials, whichhad been re-lyophilized overnight to avoid weighing errorsdue to their hygroscopic properties. All BA-PAA solutions(2.0 mg/mL) were prepared in PBS buffer at pH 7.4 to avoidpossible variations in pH.Prior to the assembly, quartz substrates (7.5 32 mm) wereetched for 30 min in piranha acid to activate the surface,rinsed with copious amounts of Milli-Q water, and dried under N2 stream. These substrates were then dipped into DABBA-PAA or ABOL-BA-PAA solution (2 mg/mL in PBS bufferpH 7.4) for 5 min, transferred into washing solution containing PBS buffer for 1 min, dipped briefly in a large amount ofMilli-Q water, transferred into ChS ARS (2 mg/mL ChSand 1 mM ARS in Milli-Q water) solution for 5 min, dippedinto the second washing solution containing Milli-Q water for1 min, and finally followed by another brief dipping in MilliQ. This cycle was repeated to reach the desired number ofbilayer. Drying under N2 stream was performed after everyBA-PAA layer deposition, excluding the very first layer. Theresulting ensemble is denoted by BA-PAA-(ChS ARS#BAPAA)n, where BA-PAA represents either DAB-BA-PAA orABOL-BA-PAA and n represents the number of bilayer.The first BA-PAA layer is regarded as a precursor layer andtherefore excluded from the bilayer number count. Typically,the ensemble consists of 5 or 10 bilayers with the poly(amidoamine) polymer as the last layer. For every multilayered system, three samples were fabricated in parallel to give estimation for standard deviation. To study the build-up profiles,UV spectra were recorded after each drying step followingBA-PAA layer formation throughout the 200–700 nm range.Every absorbance values were then corrected for baseline shiftat 700 nm. Afterwards, the multilayers were dipped intoChS ARS solution to continue multilayer build-up.Multilayers for cell culture were fabricated directly in thewells of poly-D-lysine-coated 96-well plates (PDL-TCPS,Greiner) by alternatingly dispensing deposition (70 μL) andwashing (2x 100 μL) solutions under sterile conditions insidethe laminar flow hood (LFH). Deposition started with ChS ARS (2 mg/mL ChS and 1 mM ARS in Milli-Q water,30 min for the first layer, 5 min next) as the first layer to atotal of 10 bilayers ending with the BA-PAA layer. No intermediate drying steps were applied. At the end of the fabrication process, the plates were left inside the LFH briefly to drythe films. Coated plates were kept at 4 C and used as soon aspossible (typically overnight). These multilayered samples aredesignated as PDL-(ChS ARS#BA-PAA)10 to indicate thepresence of a PDL layer as a precursor layer. Compared tomultilayers built on quartz and silicon wafer substrates, thesesystems substitute the first BA-PAA layer with PDL layer inherent to the well plate surface.ARS Release Under Physiological Conditionsand At Acidic pHARS release profiles of the three multilayered systems in PBSbuffer pH 7.4 at 37 C were investigated by dipping the thinfilms formed on quartz slides in 2 mL of PBS buffer pH 7.4solution and incubating them in a water bath with temperature set to 37 C. At regular intervals, the samples were removed, briefly dipped in a large amount of Milli-Q water,dried under N2 stream and measured by UV–vis spectrophotometer. The release study at acidic pH (pH 4, 5, and 6) wascarried out in a similar fashion, but using citrate bufferedsaline (CBS) instead of PBS.ARS Release Under Various Reductive ConditionsDegradability of the multilayered systems was investigated in asimilar way as for the investigation of their respective stabilityprofiles under physiological conditions, but in the presence of0.4 mM or 10 mM glutathione in the incubation medium.

Hujaya, Engbersen and PaulusseSolutions containing glutathione in PBS buffer at pH 7.4 wereprepared fresh directly prior to the start of experiment. Due toinstability of glutathione in PBS buffer at pH 7.4, no solutionof more than three hours old was used.ARS Release in the Presence of GlucoseThe influence of glucose concentration on ARS release profilewas investigated in a similar way as for the investigation oftheir respective stability profiles under physiological conditions, but in the presence of 25 mM and 100 mM of glucose.The 25 mM concentration is used to mimic the glucose concentration in cell culture medium. This concentration isroughly 2.5 times the minimal blood glucose level in diabeticpatients (22).were then washed with PBS, fixed with 3.7% paraformaldehyde for 15 min at RT, washed 3 times with PBS, and stainedwith Hoechst 33258 (2 μL/mL in PBS) for 15 min at RT.Samples were then washed 3 times with PBS and mountedon a clean microscope glass slide using aqueous mounting medium (Ibidi, Munich, Germany) and sealed withnail polish.Confocal microscopy was performed on an LSM 510 (CarlZeiss, Sliedrecht, The Netherlands) using Zen 2009 software.Both fluorophores (BA-ARS and Hoechst 33258) were excitedusing a 543 nm argon laser passed through HFT KP 700/S43and split through an NFT 490. BA-ARS fluorescence wasanalyzed past a BP 565–615 IR filter and Hoechst 33258excitation was analyzed past a BP 390–465 IR filter. Z-stacksectioning was performed over 20 slices throughout the33.75 μm height range.Particle Uptake by COS-7 Cells: FACSTo investigate the possibility of COS-7 cells taking up ARSboronate ester from the multilayer surface, cells were seededdirectly on multilayer-coated wells at 20 000 cells/well or 62500 cells/cm2 in complete medium with serum and left toproliferate at 37 C under humidified atmosphere with 5%CO2.After 0, 3, 6, 12, 24, and 48 h, cells were trypsinized, centrifuged (5 min, 600 g), and analyzed by FACS. Cells were alsocultured for 12 h on regular polystyrene (PS) culture plateswith and without the addition of free ARS and on multilayered films without ARS as controls to determine live cell population and fluorescence-positive cell population markers. Onsome multilayered samples treated similarly but without addition of cells, physical force was applied to break down the filmsinto the cell culture medium. The resulting suspension wasalso analyzed through FACS to locate multilayer remnantpopulation and identify their relative fluorescence intensity.Excitation of ARS-boronate ester was performed at488 nm and emission was detected via a 585 nm band-passfilter. At least 20 000 – 30 000 total events were measured toreach 10 000 events for gated living cells. Data analysis wasperformed using the FACS Cellquest Software. The gate setting was equal for all samples within the same experiment. Dotplots were applied to separate living cells population fromdead cells and film residues and particles. From the histogramobtained, markers were drawn to identify cells as live positivecells or live negative cells.Particle Uptake by COS-7 Cells: Confocal MicroscopySamples for confocal microscopy were prepared by culturingcells as described in previously. Following 24 h of culture, cellswere trypsinized and recultured inside a cover glass slide imaging chamber (General Electric Healthcare, Eindhoven,The Netherlands) for 6 h to allow the cells to attach. The cellsRESULTS AND DISCUSSIONMultilayered Thin Film Build-UpWe previously reported the syntheses of two branched boronicacid-functionalized poly(amido amine)s (BA-PAAs) (21). Thedegrees of BA functionalization of the two BA-PAAs weresimilar at 50%, with branching estimated to be 10 and 30% for DAB-BA-PAA and ABOL-BA-PAA, respectively,and number average molecular weights of 12 and 14 kg/mol,respectively. Chemical structures of the two polymers, ChS,and ARS utilized in this study are presented in Scheme 1.These polymers were employed in multilayer build-up as described in the Materials and Methods section.Figure 1 shows build-up profiles of DAB-BA-PAA-(ChS ARS#DAB-BA-PAA) 10 and ABOL-BA-PAA-(ChS ARS#ABOL-BA-PAA)10 based on increases in absorbanceat 470 nm. Incorporation of the dark red ARS drasticallychanges the spectral properties of the films, which complicatesrelating the UV spectra of these films with previous ARS-freemultilayers (21). As the ARS is the reporter molecule for theinvestigation of both loading and release, and no multilayerbuild-up was observed in the absence of ChS, it was deemedappropriate to study the film behavior based on the UVabsorption of ARS. The opaque nature of the films is veryreminiscent to the ARS-free system, indicating that ChS stillmakes up the majority of the material in the multilayeredsystem, providing thick films.Figure 1 indicates relatively linear build-up profiles for thetwo BA-PAAs. Both systems display a slow build-up in theearly stages of deposition, probably due to incomplete substrate surface coverage, causing repulsion between the substrate and the negatively-charged ARS molecules. The subsequent linear increase of ARS absorbance values indicates thatARS is incorporated systematically as the bilayer number

Poly(amido amine) Based Multilayered Thin FilmsScheme 1 Chemical structures of DAB-BA-PAA, ABOL-BA-PAA, ChS, and ARS.increases. Incorporation of ARS proved to be more efficient inthe ABOL-BA-PAA multilayers than in the DAB-BA-PAAmultilayers. This is in contrast to the previous study showingthat DAB-BA-PAA facilitated deposition of ChS and resultedin overall thicker films, as compared to ABOL-BA-PAA (21).However, this study was based on the absorbance specific toChS (at 258 nm) and therefore did not provide information onthe amount of incorporated BA-PAA polymer. The higherARS uptake in the ABOL-BA-PAA#ChS multilayers mayindicate that ABOL-BA-PAA multilayers contain a higherABOL-BA-PAA/ChS ratio than DAB-BA-PAA/ChS ratioof DAB-BA-PAA#ChS multilayers, thereby making it possible to incorporate more ARS into the films through BA-ARSinteractions. At 5 and 10 bl, ABOL-BA-PAA-based films contain 2 and 1.3 times more ARS, respectively, as compared toDAB-BA-PAA-based films at the same bl number. The insetin the UV spectra of Fig. 1 shows the digital photograph ofDAB-BA-PAA-(ChS ARS#DAB-BA-PAA)10 and ABOLBA-PAA-(ChS ARS#ABOL-BA-PAA)10. The image reveals that DAB-BA-PAA-(ChS ARS#DAB-BA-PAA)10 contains less ARS than ABOL-BA-PAA-(ChS ARS#ABOLBA-PAA)10.With DAB-BA-PAA possessing higher positive charge density due to the protonated amines in the side chains, it may beexpected that additional electrostatic interactions with thenegatively charged ARS may help to incorporate additionalARS in the DAB-BA-PAA#ChS multilayer films. However,our previous attempts to incorporate various other chargedsmall dye molecules through loading into multilayers purelyFig. 1 (a) UV-Absorption-basedbuild-up profiles (470 nm) and (b)UV spectra of DAB-BA-PAA(ChS ARS#DAB-BA-PAA)n andABOL-BA-PAA-(ChS ARS#ABOL-BA-PAA)n in the dryfilm. Inset in (a) shows the build-upprofiles of multilayers in the absenceof ARS; inset in (b) shows digitalphotograph of DAB-BA-PAA(ChS ARS#DAB-BA-PAA)10 (left)and ABOL-BA-PAA-(ChS ARS#ABOL-BA-PAA)10 (right).by electrostatic interactions were unsuccessful. Often the dyesacted as salts, enhancing deposition of polyelectrolytes withoutbeing incorporated (unpublished results). It is therefore expected that ARS-boronate ester formation is the major driving force for the successful incorporation of ARS into the (BAPAA#ChS) multilayers.As a pH indicator, ARS with a pKa of 4.5 is known tochange color from yellow to red in the pH range of 3.5 – 6.5(23). Throughout the multilayer build-up, ChS ARS deposition solution in water was red in color (pH 7), while the pHof the BA-PAA deposition solutions was maintained at 7.4through the use of phosphate buffer. If the color of ARS inthe constructs was determined solely by the pH of the deposition solutions, a red multilayered film with λmax above 500 nmwould be formed. However, the color of the films (Fig. 1, inset)was orange rather than red, with a λmax at 470 nm. This λmaxof the multilayers at 470 nm was maintained throughout the10 cycles of deposition investigated. It is known from the literature that the ARS-phenylboronate ester (BA-ARS) possesses a relatively high association constant (1300 M 1 atpH 7.4, 0.10 M phosphate buffer) (12). In contrast, ChS whichcontains vicinal diol groups with larger dihedral angles onlyforms labile boronate esters with BA, which are further weakened by electrostatic repulsion of the boronate anion by thesulfate and carboxylic acid groups of ChS. Therefore, theorange color is most likely attributed to boronate ester formation of ARS with phenylboronic acid (BA) moieties in the BAPAA (18). This finding of multilayer λmax at 470 nm is inagreement with previous findings by Ding et al. (24).

Hujaya, Engbersen and PaulusseAs further comparison, DAB-BA-PAA-(PVA ARS#DAB-BA-PAA)10 – an ARS-loaded multilayer prepared with PVA instead of ChS – remained red instead oforange (data not shown), indicating significant competitionbetween diols of PVA and ARS for ester formation with BA.Due to the multivalency and high local concentration of vicinal diol groups in PVA, the majority of BA moieties is boundto PVA, leaving the ARS molecules only physically entrappedor loosely bound in this multilayer.It is important to note that the BA-PAA#ChS multilayerfilms have an opaque appearance, which especially manifestsat higher number of layers. This hampers accurate UV determinations. Therefore, in order to minimize possible errors inUV characterization, the number of bilayers in the subsequentstudies on pH-, glucose-responsiveness, and degradation ofthe layers has been limited to 5 bilayers.ARS Release Profiles Under Physiological Conditionsand At Acidic pHDue to the pH-sensitive nature of the boronate ester formation (12), ARS is expected to be rapidly released at acidic pHwhere boronate ester formation is less favorable. To investigate the ARS release profiles from the multilayered systemsunder physiological conditions and at acidic pH, the filmswere incubated in PBS at pH 7.4, and in CBS at pH 6, 5,or 4 at 37 C. The ARS release profiles are depicted in Fig. 2indicating that within 12 h of incubation under physiologicalconditions (pH 7.4), about 30% and 40% of the incorporatedARS was released for DAB-BA-PAA and ABOL-BA-PAAsystems, respectively. ARS release continued, though substantially slower, reaching a plateau after 10 days of incubationwhere 50% of the incorporated ARS was released from themultilayered constructs. The rapid release of ARS underphysiological conditions may be facilitated by entropic gainfrom the liberation of ARS molecules before finally reachingan equilibrium at 50% ARS content.ABOL-BA-PAA multilayers, possessing a higher loading ofARS than the DAB-BA-PAA multilayers, released more of theARS content during the initial rapid release phase. Thismeans that under equal conditions and at the same numberof bilayers, much more ARS is released to the incubationmedium by the ABOL-BA-PAA system, than in the case ofthe DAB-BA-PAA system. Also, since the release was onlyfollowed by the decrease in ARS absorbance in the multilayers, it is not deducible from these results whether or notalso BA-PAA polymers or ChS were released during the initialrelease phase. Our previous results with ARS-free multilayersshowed that up to 20% of the polymeric components of themultilayers were also released within 2 h of incubation underthe same conditions (21). Therefore, the release observed inFig. 2 may represent the release of a mixture of both free ARSand ARS-boronate ester complexes.At acidic pH, ARS-boronate ester is less stable (12) andleads to more rapid ARS release. At pH 6, at least two timesfaster release of ARS is observed, as compared to the release atpH 7.4. At more acidic pH the release proceeds progressivelyfaster, and complete ARS release was achieved at pH 5 and 4,after 17 h and 5 h, respectively. In relation to the potentialapplication where these multilayers are used to coat the surface of implantable biomaterials or devices, low-pH-triggeredrelease may be exploited to deliver diol- or catecholcontaining anti-inflammatory drugs (25,26). Since implantation surgery is followed by inflammatory response, decreasingpH locally, these multilayered coatings may help in reducinginflammation by delivering the correct dosage based on thechange in pH of the environment.As discussed previously, ARS is a pH indicator. The equilibria between the different ARS species are depicted inScheme 2, including those in the presence of BA moieties. Inthe presence of these moieties, the high binding constant between ARS and BA drives the equilibrium towards boronateester formation with a distinctly different λmax. This boronateester is only stable at neutral to basic pH, where it is predominantly present in the negatively charged tetrahedral form(12). Therefore, upon decrease of pH, the equilibria areshifted towards free ARS, which are then shifted further tothe yellow species (λmax 419 nm) by the low pH. Figure 3shows the expected decrease in λmax during the incubation inthe different acidic pH values. As shown in Scheme 2, this shiftin λmax is accompanied by ARS-boronate ester hydrolysis,leading to ARS release from the films.During the first hour of exposure to PBS buffer at pH 7.4,the λmax of the multilayers remains stable at 470 nm, followedby a shift to longer wavelength throughout the rest of theincubation period. The shift in λmax reached a maximumvalue at 510 nm after about 20 days of incubation. Unlikein the case of acidic pH, this shift in λmax progressed muchmore slowly, underlining the high binding affinity of the BA toARS at pH 7.4. The initial 1 h rapid release of ARS (Fig. 2) isnot accompanied by a shift in λmax. This further indicates mostof the incorporated ARS is bound to BA moieties or involvedin dynamic ester formation (27) with the BA moieties in thefilm. In the later stages of incubation, slow gradual boronateester hydrolysis and diffusion of ARS take place, leavinghigher portions of unbound physically-entrapped ARS in themultilayer, resulting in a shift of λmax to that corresponding tofree ARS at pH 7.4 (λmax 519 nm).ARS Release Under Reducing ConditionsOwing to the presence of disulfide bonds in the main polymerchain of the BA-PAAs, the resulting multilayered ensemblesare responsive to the presence of reducing agents as well. Thisreducibility is demonstrated by exposing the multilayers to0.4 mM and 10 mM of glutathione in the incubation solutions

Poly(amido amine) Based Multilayered Thin FilmsFig. 2 ARS release profiles underphysiological conditions and acidicpH in vitro. Left part of the abscissashows release within 1 day; rightpart shows slower release duringremaining incubation.release and film degradation was achieved after 10 h ofincubation.of PBS at pH 7.4. Figure 4 shows the ARS release profilesunder these reducing conditions. In the presence of 0.4 mMglutathione, ARS release is not significantly enhanced as compared to that observed in the absence of glutathione, at leastduring the first 10 h of incubation. This slow response to0.4 mM glutathione was also observed for the degradationof the ARS-free multilayers, and can be attributed to theirrelatively high structural stability due to the branched structure of the polymers. However, during prolonged incubation,disulfide reduction progressed more extensively (incubationduration 10 h), ARS release displayed an increased releasecompared to the release observed in the absence of glutathione. At the higher glutathione concentration of 1

ChS ARS solution to continue multilayer build-up. Multilayers for cell culture were fabricated directly in the wells of poly-D-lysine-coated 96-well plates (PDL-TCPS, Greiner) by alternatingly dispensing deposition (70 μL) and washing (2x 100 μL) solutions under sterile conditions inside the laminar flow hood (LFH). Deposition started with ChS