Article pubs.acs.org/Langmuir

Two Methods for One-Point Anchoring of a Linear Polysaccharide on a Gold Surface Julia Hoypierres, Virginie Dulong, Christophe Rihouey, Stéphane Alexandre, Luc Picton, and Pascal Thébault* Rouen Université, Normandie, France Laboratoire Polymères Biopolymères Surfaces, Université de Rouen, F-76821 Mont Saint Aignan, France CNRS UMR 6270 & FR3038, F-76821 Mont Saint Aignan, France S Supporting Information *

ABSTRACT: Two strategies to achieve a one-point anchoring of a hydrolyzed pullulan (P9000) on a gold surface are compared. The first strategy consists of forming a selfassembled monolayer of a 6-amino-1-hexanethiol (AHT) and then achieving reductive amination on the surface between the aminated surface and the aldehyde of the polysaccharide reductive end sugar. The second consists of incorporating a thiol function at the extremity of the pullulan (via the same reductive amination), leading to P9000-AHT and then immobilizing it on gold by a spontaneous reaction between solid gold and thiol. The modified pullulan was characterized by NMR and size-exclusion chromatography coupled to a light-scattering detector. P9000-AHT appears to be in a disulfide dimer form in solution but recovers its unimer form with dithiothreitol (DTT) treatment. The comparison of the two strategies by contact angle and XPS revealed that the second strategy is more efficient for the pullulan one-point anchoring. P9000-AHT even in its dimer form is easily grafted onto the surface. The grafted polymer seems to be more in a coil conformation than in a rigid brush. Furthermore, QCM measurements highlighted that the second strategy leads to a grafting density of around 3.5 × 1013 molecules·cm−2 corresponding to a high surface coverage. The elaboration of a dense and oriented layer of polysaccharides covalently linked to a gold surface might enhance the use of such modified polysaccharides in various fields.

■

or layer-by-layer deposition6,7) or by covalent grafting. However, a majority of covalent coatings are made by using the lateral functions of the polysaccharide chains, named multipoint anchoring. Only a few authors made polysaccharide coatings by using only one function at the terminal reducing part. This is the case for Nordgren et al.,8 who incorporate a thiol function at the extremity of a xyloglucan through an enzymatic method in order to anchor it by a single point on a gold surface. However, this technique is specific to xyloglucan and cannot be applied to other polysaccharides.9 Oligosaccharides or monosaccharides are widely used in the field of biosensors, particularly for the design of carbohydrate arrays.10 These devices are based on carbohydrate recognition (largely present in nature) and are used to investigate, inter alia, protein−carbohydrate interactions and cellular recognition.11 A majority of them are composed of a substrate such as glass,12 silicon,13 or gold14 onto which carbohydrates are covalently

INTRODUCTION In the last few decades, the control of surface chemistry and topography allowed the development of various innovative technologies, such as super hydrophobic/hydrophilic surfaces, antibacterial surfaces, and biosensors. Moreover, in the field of green chemistry, natural molecules for surface modification are increasingly used, in particular, carbohydrates such as polysaccharides, oligosaccharides, disaccharides, and monosaccharides. The grafting of such molecules on surfaces enables their design with particular properties. Thus, some authors designed bacteria-repellent PVC surfaces,1 antiadhesive titanium surfaces,2 or antibacterial polyurethane surfaces3 by grafting different polysaccharides on the substrates. Polysaccharides are used in other fields, such as packaging. For example, Introzzi et al.4 elaborated an antifog coating based on pullulan for food packaging. In a biomedical field, a coating based on alginic acid or chondroitin on an artificial surface enables a significant reduction of the thrombogenicity of this surface.5 Usually, coatings are made either by a simple deposition on the surface (e.g., spin coating © 2014 American Chemical Society

Received: October 24, 2014 Revised: December 11, 2014 Published: December 14, 2014 254

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Article

Langmuir immobilized. Seo et al. 15 immobilized three types of carbohydrates on a gold surface, i.e., glucose (a monosaccharide), lactose (a disaccharide), and a GM1 pentasaccharide. They used aminophenyl disulfide to incorporate a thiol function at the free reducing end of the sugars through a reductive amination reaction. The thiol function will react spontaneously with solid gold and form a sulfur−gold bond.16 This technique allows control of the carbohydrate immobilization and enables the grafting on the surface by a single point at its terminal reducing part. Zhi et al.17 investigated two different strategies to immobilize oligosaccharides (composed of 5 or 10 sugars) on a gold surface through their free reducing ends. The first strategy consists of the chemioselective attachment of a nonderivatized oligosaccharide, and the second one consists of immobilizing an end-derivatized oligosaccharide on a functionalized gold surface. However, to our knowledge no study proposes to graft polysaccharides (composed of more than 20 sugars) on gold surfaces by the use of their terminal reducing part. Here, we propose to graft a linear polysaccharide (pullulan) on a gold surface through its free reducing end. Two strategies are studied (Figure 1): the first by incorporating amine

mix between solution and surface chemistry, the main drawback should be the purification steps of the modified polysaccharide (time consumer of dialysis, elimination of side reaction byproducts). However, the advantages for this second strategy are the easy characterization of modified polysaccharide in solution and probably a better kinetic and control of the polysaccharide immobilization on the surface because the thiol−gold reaction is expected to be more favorable than the reductive amination on the surface (first strategy). The two strategies were compared by contact angle, X-ray photoelectron spectroscopy (XPS), and quartz-crystal microbalance (QCM) analyses.

■

EXPERIMENTAL SECTION

Materials. Pullulan used in this study (Mn = 125 000 g·mol−1 and Mw = 265 000 g·mol−1, determined by SEC/MALS/dRI) was purchased from Hayashibara Biochemical Laboratory (Okayama, Japan). Hydrochloric acid (HCl), sodium hydroxide (NaOH), 6amino-1-hexanethiol hydrochloride (AHT), sodium cyanoborohydride (NaBH3CN), dimethyl sulfoxide (DMSO), absolute ethanol, and deuterium oxide (D2O) were purchased from Sigma-Aldrich. Dithiothreitol (DTT) was purchased from Bio-Rad Laboratories, Inc. All aqueous solutions were prepared with Milli-Q water. Dialysis tubing (Spectrapore, cutoff points 1000 and 3000 g·mol−1) was purchased from Spectrum Laboratories. The gold surfaces are borosilicate glass substrates coated with chromium and gold layers of 2.5 ± 1.5 and 250 ± 50 nm thickness, respectively. The substrates are 11 × 11 mm2 in size and were purchased from Arrandee (Germany). Polymer Chemical Modification. Pullulan Hydrolysis. An aqueous solution of precursor pullulan was prepared in Milli-Q water at 20 g·L−1, and the pH was adjusted to 2 with 4 N HCl. The solution was held at 80 °C and placed under reflux and magnetic stirring for 72 h (hydrolysis time determined via a kinetic study). Then the solution was cooled to room temperature, and the pH was neutralized with 1 N NaOH. Finally, the solution was dialyzed against Milli-Q water (cutoff point 3000 g·mol−1) and freeze dried. The final product was characterized by NMR and SEC/MALS/dRI. The hydrolyzed pullulan was labeled as P9000, regarding the obtained Mn of 9000 g mol−1 (discussed in the following text). Hydrolyzed Pullulan Thiolation. One gram of P9000 (1 mol equiv) was dissolved in a solution containing 235.5 mg of NaBH3CN (34 mol equiv) and 215 mg of AHT (11 mol equiv) in 50 mL of DMSO. The solution was held at 60 °C for 72 h under reflux and magnetic stirring. The solution was then cooled at room temperature and purified by precipitation in acetone. The resulting solid was dissolved in Milli-Q water and dialyzed against a mixture of ethanol and Milli-Q water (50/ 50) for 3 days and against Milli-Q water for 7 days (cutoff point 1000 g·mol−1). The product was finally freeze dried. The final product was characterized by NMR and SEC/MALS/VD/dRI. Such modified P9000 was labeled as P9000-AHT. Polymer Characterization. SEC/MALS/VD/dRI Analysis. Weightand number-average molar masses (Mw and Mn), polydispersity indexes, and the hydrodynamic radius (Rh) of pullulan, hydrolyzed pullulan, and thiolated pullulan were determined by size exclusion chromatography (SEC) coupled inline with multiangle light scattering (MALS), a viscosity detector (VD), and a refractive index (dRI) detector. The MALS detector is a DAWN Heleos-II (Wyatt Technology Inc., USA), the viscometer is a ViscoStar (Wyatt Technology. Inc., USA), and the dRI detector is a RID-10A (Shimadzu, Japan). The columns were OHPAK SB 802.5 HQ and OHPAK SB 804 HQ columns from Shodex (USA). The eluted solvent (0.1 M LiNO3) was filtered through a 0.1 μm filter unit (Millipore, USA) and degassed online (Shimadzu DGU-20A3R, Japan). The polysaccharide solutions (7.5 g·L−1 for pullulan and 15 g·L−1 for hydrolyzed pullulan and thiolated pullulan) were prepared in the eluent and filtered on a 0.45 μm unit filter (Interchim, France) The experimentally determined dn/dC values were 0.147 and 0.135 L·g−1

Figure 1. Schematic representation of the two strategies employed for the P9000 immobilization on gold surfaces.

functions on the gold surface and then by grafting the pullulan through a reductive amination reaction; the second by modifying the aldehyde reducing end of the polysaccharide via the introduction of a thiol function through a reductive amination and then by immobilizing the modified compound directly on the gold surface. The two strategies are expected to present advantages and disadvantages. Indeed, the first strategy requires very small amounts of reagents because it consists only of a chemical modification of the surface. Furthermore, “purification” steps after the surface chemical modification are generally conducted by simple rinsing procedures in an appropriate solution. As a possible disadvantage, the fixation of the polysaccharide on the aminated surface via its reductive extremity should be more difficult than an equivalent reaction conducted in solution (strategy 2), especially because the length of the polymer chain is not negligible in our case. Consequently, the pullulan immobilization (second step of the first strategy) should be longer (or have a lower yield) than the equivalent functionalization of the pullulan in solution (first step of the second strategy). Concerning the second strategy, which is a 255

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Langmuir

■

for the precursor and hydrolyzed pullulan, respectively. The collected data were analyzed using the Astra 6 software package from Wyatt Technology Inc. The Rh was determined according to the measurement of the intrinsic viscosity using the Einstein−Simha equation: Vh = [η]M/νNA, where Vh is the hydrodynamic radius (Vh = Rh3π4/3), NA is Avogadro’s number, M is the molar mass, [η] is the intrinsic viscosity (g·mL−1), and ν is a conformational parameter that takes a value of 2.5 in the case of a spherical conformation, which was expected because of the high flexibility of pullulan (random coil). NMR Analysis. NMR experiments were carried out on Bruker Advance (300 and 500 MHz) spectrometers at a concentration of 50 g·L−1 in D2O. The acquisitions occurred at room temperature (25 °C). The chemical shifts are in parts per million (ppm). The processing of NMR data was performed with a 1D NMR processor from ACD Laboratories. Surface Modification. Gold Surface Preparation. Gold surfaces were first annealed in a butane flame and then cleaned twice in absolute ethanol baths under sonication (5 min). The substrates were then dried under a flow of dry nitrogen. First Strategy. Cleaned gold surfaces were immersed in a solution of AHT at 5 mM in absolute ethanol. The solution was stirred on an orbital shaker for 24 h. The substrate was then rinsed under a flow of ethanol, and nonimmobilized molecules were removed thanks to two absolute ethanol baths under sonication (5 min). Finally, the substrate was dried under a flow of dry nitrogen. The gold surface previously modified by aminothiol (Au-AHT) was immersed in 10 mL of a solution of hydrolyzed pullulan (P9000) at 1 g·L−1 in DMSO and 34 mol equiv of NaBH3CN (2.35 mg). The solution was stirred on an orbital shaker at 25 °C for 7 days (which is the optimal reaction time tested). The surface was then rinsed with Milli-Q water and dried under a flow of dry nitrogen. Second Strategy. Forty milligrams of P9000-AHT was dissolved in 10 mL of Milli-Q water for 16 h, and a cleaned gold surface was immersed for 3 h at room temperature. The surface was then rinsed under a flow of Milli-Q water and dried under dry nitrogen. Surface Characterization. Contact Angle. Static contact angle measurements with Milli-Q water were carried out with a Multiskop goniometer (Optrel GmbH) equipped with a CDD camera and a syringe (equipped with a micrometric screw-type piston). Water angle values were the average of three measurements (2.5 μL drops) on three different samples each. Contact angles were calculated with the CAM software. X-ray Photoelectron Spectroscopy Characterization (XPS). XPS measurements were carried out with a Thermo Electron K-Alpha spectrometer using a monochromatic Al Kα X-ray source (1486.6 eV) with a spot size of 400 μm. The survey spectra were collected over a range of −10 to 1350 eV with a pass energy of 200 eV. The highresolution spectra over the C 1s, O 1s, S 2p, and N 1s regions were acquired with a pass energy of 50 eV. The binding energy was referenced by setting the maximum C 1s peak at 285.0 eV. The deconvolution of peaks and the determination of the proportion of component composition from peak areas were performed using a CasaXPS software system. Quartz Crystal Microbalance Tests (QCM). The immobilization of P9000-AHT on a gold-coated quartz crystal was monitored by quartz crystal microbalance (QCM) measurements using a D300 system (QSense AB, Sweden) with a QAFC 302 flow chamber. This technique consists of measuring the resonance frequency shift Δf of a quartz crystal sensor induced by material deposition onto the surface. Changes in the resonance frequencies were measured at the third overtone (n = 3), corresponding to the 15 MHz resonance frequency. A shift in Δf can be associated, in a first approximation, to a variation of the mass adsorbed onto the crystal through the Sauerbrey relation: m = CΔf/n, where C is a constant characteristic of the crystal used (C = 17.7 ng·cm−2·Hz−1). The processing consists of introducing a 2 g· L−1 aqueous solution of P9000-AHT in the system and following the frequency variations for 140 min. When the frequency was stabilized, the measuring cell was then rinsed with a flow of Milli-Q water in order to eliminate the nonimmobilized molecules.

Article

RESULTS AND DISCUSSION Chemical Modification. Pullulan Hydrolysis. Pullulan hydrolysis is carried out in order to obtain shorter polysaccharide chains (smaller average molar masses). This hydrolysis could allow us to have better control of the pullulan grafting on the surface. Several methods exist for the degradation of pullulan. In a preliminary study, we have first tried to use an enzyme, the pullulanase, that is able to cleave α1,6 boundaries of pullulan.18 However, the obtained hydrolyzed pullulan presented too high a polydispersity index (Đ = 2.4), which is not suitable for our study. Moreover, this enzyme is not able to hydrolyze a chemically modified pullulan (such as carboxymethylpullulan19), which could be used for future studies. Thus, we decided to use acid hydrolysis by working in hydrochloric acid media. We followed the kinetic of pullulan hydrolysis (20 g·L−1) at pH 2 and 80 °C (see Supporting Information Figures SI 1,2). On the basis of these results, the time was fixed at 72 h. The obtained polymer after dialysis was characterized by SEC/MALS/dRI, and it presents a numberaverage molar mass of 9 × 103 g mol−1 and a weight-average molar mass of 11.5 × 103 g mol−1, with a good polydispersity index of 1.3. This oligosaccharide was then noted as P9000. As shown in Figure 2, the 13C NMR spectra of pullulan and

Figure 2. 13C NMR spectra of (a) precursor pullulan and (b) acidhydrolyzed pullulan.

hydrolyzed pullulan are very similar. Moreover, no additional peak corresponding to the formation of supplementary functions such as aldehyde is observed. This signifies that acid hydrolysis on pullulan did not alter the structure of the polysaccharide, regarding the apparatus sensitivity, and it will allow us to control the pullulan grafting on the surfaces. P9000 Thiolation. The P9000 extremity was modified by AHT in order to incorporate a thiol function on the pullulan chain. The amine function of the aminothiol compound reacted with the unique aldehyde of the P9000 through a reductive amination reaction.20 This chemically modified product was named P9000-AHT. The 1H NMR spectrum of P9000-AHT (Figure 3) confirms the reaction success. All peaks for pullulan 256

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Article

Langmuir

Figure 3. 1H NMR spectrum (300 MHz) of P9000-AHT.

are present, with those at 5.38 and 4.95 ppm corresponding to the anomeric protons of pullulan. Furthermore, new peaks appeared corresponding to the presence of AHT. The peaks at around 1.71 and 1.42 ppm are assigned to aminothiol protons situated at positions b, e and c, d of the aminothiol alkyl chain, respectively, and the peak around 3.10 ppm is assigned to protons situated at position a. (Figure 3). The signals at 1.07 and 1.22 ppm are probably due to impurities. Finally, the peak at 2.76 ppm corresponds to protons situated α to a disulfide bond21 (position f) and not α to a thiol function (usually around 2.5 ppm22). From this observation, it is possible to deduce that the majority of products are in disulfide forms and not thiol, as expected. Thus, on the basis of qualitative and quantitative (through integration) analysis of the 1H NMR spectrum of P9000-AHT, the final product is probably a mix of four compounds: free P9000, which did not react; P9000-NH(CH2)6-SH (encoded PS) in the thiol form (probably a small amount because none was detectable by 1H NMR); and disulfide forms P9000-NH-(CH2)6-S-S-(CH2)6-NH-P9000 (encoded as PSSP) and P9000-NH-(CH2)6-S-S-(CH2)6-NH2 (encoded as PSS). Unfortunately, this mixture of compounds and the weak signal from protons of grafted AHT compared to the strong signal from protons of P9000 lead to an imprecise integration and therefore do not allow an estimation of the grafting rate. The product was also characterized by SEC/MALS/dRI (Figure 4). Mn and Mw were respectively 17 × 103 and 22 × 103 g·mol−1. These values are approximately twice as high as the molar masses of precursor P9000 (9 × 103 g mol−1), which allowed us to assume that a great majority of the compounds are in the PSSP disulfide form. This also indicates that unreacted P9000 and PSS are not present (or are present in very small amounts) in the final product. The PSSP was therefore treated with excess dithiothreitol (DTT) solution, which is able to reduce disulfides to thiols.23 Indeed after such treatment, Mn and Mw values are 11 × 103 and 13 × 103 g· mol−1, respectively (Figure 4). These results are close to the P9000 (or PS) molar masses and confirm that P9000-AHT without DTT (i.e. at the end of the reaction) was the major component under the disulfide form. Moreover, if we compare the two chromatograms before and after DTT treatment (Figure 4), then it is noticeable that, for the same elution volume (and thus the same hydrodynamic volume), the disulfide presents a molar mass that is the double that of the thiol molar mass. This means that the disulfide is 2 times more

Figure 4. Elution profiles obtained by SEC/MALS/DRI from the refractive index (dotted lines) and LS (full lines) of P9000-AHT (gray) and P9000-AHT treated for 2 h with DTT (black) together with the molar mass distribution in 0.1 M LiNO3.

dense than the thiol. This result may be surprising but can be explained by intramolecular interactions between the two hydrocarbon chains of the disulfide leading to a hairpin conformation. Finally, DTT is a useful tool for breaking disulfide bonds, and for further P9000-AHT synthesis, it will be used during dialysis in order to convert the PSS (even if it is present in a very small amount in the final product) into PS and to eliminate residual AHT. To confirm the above conclusion (i.e., the majority of the product reaction is PSSP), we have looked at other methods of quantifying thiol (colorimetric assay with Ellman’s reagent), free P9000 (reductive sugar titration via colorimetric assays using bicinchoninic acid or 3,5-dinitrosalicylic acid), and PSS (primary amine titration via a fluorometric assay using ophtalaldehyde in the presence of mercaptoethanol24 or spectrometric assays using 2,4,6-trinitrobenzenesulfonic acid25). Ellman’s test revealed that PS was present at around 2%, which is a very small ratio, confirming the previous conclusions. However, even if the amount of thiol function is small, its presence is sufficient to prevent the use of both 257

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Article

Langmuir described colorimetric and fluorometric methods for the quantification of primary amines and reductive sugars.24−27 Surface Modification. P9000 was grafted on gold surfaces following two strategies. The first involves forming a selfassembled monolayer (SAM) with the AHT28 and then by grafting the P9000 on the amine functions on the surface through its reductive extremity. The second is to immobilize the P9000-AHT compound on gold by a spontaneous reaction between solid gold and sulfur. The two strategies were compared by contact angle measurements, XPS analysis, and AFM images. Contact Angle. Contact angle measurements were carried out in order to compare the hydrophilicity of the different surfaces. Figure 5 shows the contact angle values of cleaned

Table 1. Surface Atomic Composition of Modified Gold, as Determined by XPS surface

% Au 4f

%C 1s

%O 1s

%N 1s

%S 2p

thickness (Å)

cleaned gold AHT first strategy second strategy

54.4 42.9 31.8 19.3

26.0 29.7 39.5 44.4

19.6 21.4 24.7 34.1

0 2.6 2.1 0.8

0 3.3 1.8 1.4

4 12 30

are different than the theoretical ratio (i.e., N/S = 1). This is probably due to the low signal of both atoms. Concerning the second strategy, the presence of these atoms on the modified surface seems to attest to the grafting of P9000-AHT. Furthermore, the carbon and oxygen rates increase for the modified surfaces, whatever the strategy used. It confirms the presence of pullulan on the surfaces. However, the increase in both carbon and oxygen atoms together with the diminishing of gold is amplified in the case of the second strategy. The layer thickness of modified gold surfaces was estimated by XPS using the following equation ⎛ d ⎞ ⎟ = exp⎜ − ⎝ λ sin θ ⎠ A(Au4f 7/2) A(Au4f 7/2)

0

where d is the layer average thickness, θ is the takeoff angle fixed at 90° in our case, and λ is the escape depth of Au 4f7/2 electrons through an organic layer, calculated to be 36 Å.29 For the gold surface modified only by a SAM of AHT, the layer thickness is around 4 Å, whereas it is around 12 Å for the first strategy and almost 30 Å for the second strategy. This should reflect a higher quantity of grafted P9000 on the gold surface with the second strategy. The P9000 chain length was estimated to be 39 nm (with a Mn of 9000 g·mol−1 and a Kuhn segment length of 2.4 nm),30 10 times greater than the measured thickness. This indicates that the surface-grafted pullulan does not form a rigid brush. Regarding the hydrodynamic radius of the P9000 random coil in aqueous solution, which was estimated by SEC/MALS/VD/dRI to be 2.2 nm, it seems more consistent to say that P9000 is in the form of coils on the surface, although this value is obtained in aqueous solution whereas the XPS analyses were carried out in a dry state. The XPS C 1s spectra of the modified gold surfaces by the two strategies are represented in Figure 6. For both strategies, the C 1s peak was best fit with four contributions. The first one, at the lowest binding energy (BE) of 285.0 eV, is attributed to carbon in C−C bonds, and the second at around 286.7 eV represents the carbon atoms in C−O bonds (or C−N or C−S). The third contribution, at around 288.2 eV, is assigned to C atoms involved in O−C−O bonds. Finally, the last one at around 289.5 eV probably corresponds to impurities. These features are consistent with other studies.31,32 The major differences between the two spectra are the quantitative chemical composition (Table 2). Indeed, for the first strategy, the most intensive signal corresponds to carbon in C−C bonds, whereas for the second strategy, the main signal corresponds to carbon at the α position of a heteroatom such oxygen (C−O), nitrogen (C−N), or sulfur (C−S). On the pullulan backbone, all carbon atoms are linked to at least one oxygen atom. Consequently, the XPS C 1s spectrum of the modified gold surface via the second strategy seems to be more consistent with the pullulan structure. Therefore, we can deduce that the second strategy is more effective for the P9000

Figure 5. Water contact angle of the cleaned gold surface and modified gold surfaces. Error bars represents standard deviations.

gold surface and gold surfaces modified by AHT and by P9000 through the two strategies. The contact angle value decreases after AHT immobilization. This phenomenon can be explained by the presence of amine functions, which are in the hydrophilic ammonium form in the presence of Milli-Q water. Concerning the P9000-modified gold surfaces by the two strategies, contact angle values are different from those of the AHT-modified gold surface. However, the contact angles are strongly different regarding the grafting strategy (60 and 30° for strategies 1 and 2, respectively). This result indicates that the grafting (i.e., the amount and/or the organization of grafted pullulane on the surface) differs according to the choice of the strategy. Typically, the immobilization of polysaccharides leads to highly hydrophilic surfaces.4 Then, the second strategy seems to be the better one. The result obtained for the first strategy is surprising (i.e., increase in contact angle). We could think that the long time (7 days) and conditions (DMSO and NaBH3CN) of the reaction for this strategy might affect the surface. This also constitutes an argument against the first strategy. XPS. XPS analyses were carried out on a cleaned gold surface, AHT layer, and gold surfaces modified by the two strategies to provide chemical information on the different modifications. As shown in Table 1, the appearance of nitrogen and sulfur atoms confirms the formation of the AHT layer. However, the experimental ratios between nitrogen and sulfur 258

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Article

Langmuir

Figure 6. XPS C 1s spectra of P9000-modified gold surfaces obtained by (a) the first strategy and (b) the second strategy.

XPS S 2p spectrum of modified gold surfaces by the first strategy. It shows two peaks at 162.1 and 163.3 eV corresponding to sulfur in S−Au bonds (i.e., S 2p3/2 and S 2p1/2, respectively) and two peaks corresponding to free thiol or disulfide at 164.2 and 165.4 eV. The proportions of S−Au bonds and free thiol are 69.6 and 30.4%, respectively. The S− Au signal confirms the formation of the SAM of AHT and shows its persistence even after pullulan grafting. The presence of free thiol means that some AHT remains physically adsorbed on the surface. Figure 7b shows the XPS S 2p spectrum of modified gold surfaces by the second strategy. As the previous spectrum, it shows a pair of peaks at 162.5 and 163.7 eV corresponding to sulfur in S−Au bonds (29.7%) and another pair at 163.9 and 165.1 eV corresponding to free thiol or disulfide (31.3%). The presence of the S−Au signal indicates that P9000 is covalently linked to the gold surface, and the presence of free thiol or disulfide means that some polymers remain adsorbed on the surface. Furthermore, a third pair of

Table 2. C 1s Chemical Compositions (Atom %) by XPS of P9000-Modified Gold Surfaces surface first strategy

second strategy

chemical composition

atom %

C−C (285.0 eV) C−O, C−N, C−S (286.8 eV) O−C−O (288.4 eV) impurity (289.4 eV) C−C (284.9 eV) C−O, C−N, C−S (286.6 eV) O−C−O (288.0 eV) impurity (289.6 eV)

66.9 25.9 4.1 3.1 37.2 45.0 15.5 2.3

grafting. The C−C signal for the first strategy is probably the most intensive because of residual unreacted AHT molecules in the first layer. The covalent binding through Au−S bonds by both methods was investigated by the XPS S 2p analysis. Figure 7a shows the

Figure 7. XPS S 2p spectra of P9000-modified gold surfaces obtained by (a) the first strategy and (b) the second strategy. 259

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Langmuir

■

peaks is also observed at 161.4 and 162.6 eV, with a proportion of 39%. These peaks have already been observed in the literature and attributed either to atomic sulfur33 or to polycoordinated sulfur.34,35 Wirde and Gelius36 have explained that this sulfur polycoordination on gold often occurs with the immobilization of disulfides because of the proximity of the two sulfur atoms on the gold surface. This explanation is consistent with our results. Indeed, this peak is observed only on the XPS S 2p spectrum of the second strategy, which involved P9000AHT compounds in the disulfides form as observed by SEC/ MALS/VD/dRI and 1H NMR. This signal also confirms the P9000 grafting on the gold surface. QCM. The immobilization of P9000-AHT on gold (by using the second strategy) was characterized by QCM in order to estimate the quantity of immobilized molecules. This experiment was carried out only with the second strategy because the first strategy requires the use of a hazardous reactant and solvent that could alter and contaminate the apparatus. A low variation of dissipation was observed, allowing us to link the frequency and mass variation through the Sauerbrey equation. The measured −Δf/3 was equal to 30 Hz, which corresponds to 530 ng·cm−2 P9000-AHT deposited on the quartz crystal. This value corresponds to a grafting density of around 3.5 × 1013 molecules·cm−2, which is similar to the results obtained by Nordgren et al.,8 whose work was described previously. These results confirm that the chemisorption of P9000-AHT occurred with a high surface coverage.

Article

ASSOCIATED CONTENT

S Supporting Information *

Pullulan acid hydrolysis monitoring and QCM surface characterization. This material is available free of charge via the Internet http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Funding

We acknowledge financial support from the A-I Chem Channel project that was selected by the European INTERREG IV A France (Channel)−England Cross-Border Cooperation Program. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Lafarge, J.; Kébir, N.; Schapman, D.; Gadenne, V.; Burel, F. Design of bacteria repellent PVC surfaces using the click chemistry. Cellulose 2013, 20, 2779−2790. (2) Gadenne, V.; Lebrun, L.; Jouenne, T.; Thebault, P. Antiadhesive activity of ulvan polysaccharides covalently immobilized onto titanium surface. Colloids Surf., B 2013, 112, 229−236. (3) Kara, F.; Aksoy, E. A.; Yuksekdag, Z.; Hasirci, N.; Aksoy, S. Synthesis and surface modification of polyurethanes with chitosan for antibacterial properties. Carbohydr. Polym. 2014, 112, 39−47. (4) Introzzi, L.; Fuentes-Alventosa, J. M.; Cozzolino, C. A.; Trabattoni, S.; Tavazzi, S.; Bianchi, L.; Schiraldi, A.; Piergiovanni, L.; Farris, S. Wetting Enhancer Pullulan Coating for Antifog Packaging Applications. ACS Appl. Mater. Interfaces 2012, 4, 3692−3700. (5) Keuren, J. F. W.; Wielders, S. J. H.; Willems, G. M.; Morrac, M.; Cahaland, L.; Cahaland, P.; Lindhout, T. Thrombogenicity of polysaccharide-coated surfaces. Biomaterials 2003, 24, 1917−1924. (6) Kadi, S.; Cui, D.; Bayma, E.; Boudou, T.; Nicolas, C.; Glinel, K.; Picart, C.; Rachel Auzély-Velty, R. Alkylamino Hydrazide Derivatives of Hyaluronic Acid: Synthesis, Characterization in Semidilute Aqueous Solutions, and Assembly into Thin Multilayer Films. Biomacromolecules 2009, 10, 2875−2884. (7) Pozos Vázquez, C.; Boudou, T.; Dulong, V.; Nicolas, C.; Picart, C.; Glinel, K. Variation of Polyelectrolyte Film Stiffness by PhotoCross-Linking: A New Way To Control Cell Adhesion. Langmuir 2009, 25, 3556−3563. (8) Nordgren, N.; Eklöf, J.; Zhou, Q.; Brumer, H. III; Rutland, M. W. Top-Down Grafting of Xyloglucan to Gold Monitored by QCM-D and AFM: Enzymatic Activity and Interactions with Cellulose. Biomacromolecules 2008, 9, 942−948. (9) Brumer, H. III; Zhou, Q.; Baumann, M. J.; Carlsson, K.; Teeri, T. T. Activation of Crystalline Cellulose Surfaces through the Chemoenzymatic Modification of Xyloglucan. J. Am. Chem. Soc. 2004, 126, 5715−5721. (10) Park, S.; Gildersleeve, J. C.; Blixt, O.; Shin, I. Carbohydrate microarrays. Chem. Soc. Rev. 2013, 42, 4310−4326. (11) Zeng, X.; Andrade, C. A. S.; Oliveira, M. D. L.; Sun, X. L. Carbohydrate−protein interactions and their biosensing applications. Anal. Bioanal. Chem. 2012, 402, 3161−3176. (12) Zhou, X.; Zhou, J. Oligosaccharide microarrays fabricated on aminooxyacetyl functionalized glass surface for characterization of carbohydrate−protein interaction. Biosens. Bioelectron. 2006, 21, 1451−1458. (13) Leone, G.; Consumi, M.; Lamponi, S.; Magnani, A. Combination of static time of flight secondary ion mass spectrometry and infrared reflection-adsorption spectroscopy for the characterisation of a four steps built-up carbohydrate array. Appl. Surf. Sci. 2012, 258, 6302−6315.

■

CONCLUSIONS We compared two methods to carry out a one-point anchoring of a hydrolyzed polysaccharide on a gold surface through its extremity. The first is conducted by forming a SAM of AHT and then carrying out a reductive amination reaction between the aminated surface and the end reductive sugar (aldehyde) of the hydrolyzed pullulan (P9000). The second is conducted by incorporating a thiol function (via reductive amination) at the terminal reducing part of the pullulan (P9000-AHT) and then immobilizing it on gold by a spontaneous reaction between the surface and sulfur atom. The P9000-AHT synthesis was characterized qualitatively by 1 H NMR and SEC/MALS/dRI, revealing that the major part of the product is under the disulfide form PSSP. The two anchoring strategies were compared by contact angle and XPS. These techniques indicate that the second strategy allows a better grafting of pullulan on the gold surface. Indeed, the low availability of amine functions when they are grafted on the surface leads to a low reaction yield. Furthermore, QCM measurements show a high coverage of P9000 on the surface through the second strategy, confirming P9000 grafting on a more regular basis. The control of the covalent grafting of such macromolecular molecules could lead to an increase in their use, in particular, in the biosensor field. Chemical modifications of pullulan are under consideration to introduce new properties onto the gold surface and therefore new applications for this system. Experiments to evaluate the binding of proteins, such as bovine serum albumin (BSA) or maltose-binding protein (MPB),37 for example, will be realized on gold surfaces modified by pullulan and derivatives (anionic, amphiphilic,...) to study parameters involved in interactions between proteins and polysaccharides. 260

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261

Article

Langmuir (14) Zhi, Z. L.; Powell, A. K.; Turnbull, J. E. Fabrication of Carbohydrate Microarrays on Gold Surfaces: Direct Attachment of Nonderivatized Oligosaccharides to Hydrazide-Modified Self-Assembled Monolayers. Anal. Chem. 2006, 78, 4786−4793. (15) Seo, J. H.; Adachi, K.; Lee, B. K.; Kang, D. G.; Kim, Y. K.; Kim, K. R.; Lee, H. Y.; Kawai, T.; Cha, H. J. Facile and Rapid Direct Gold Surface Immobilization with Controlled Orientation for Carbohydrates. Bioconjugate Chem. 2007, 18, 2197−2201. (16) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-Assembled Monolayers of Thiolates on Metals as a Form of Nanotechnology. Chem. Rev. 2005, 105, 1103−1170. (17) Zhi, Z. L.; Laurent, N.; Powell, A. K.; Karamanska, R.; Fais, M.; Voglmeir, J.; Wright, A.; Blackburn, J. M.; Crocker, P. R.; Russell, D. A.; Flitsch, S.; Field, R. A.; Turnbull, J. E. A Versatile Gold Surface Approach for Fabrication and Interrogation of Glycoarrays. ChemBioChem 2008, 9, 1568−1575. (18) Spreinat, A.; Antranikian, G. Purification and properties of a thermostable pullulanase from Clostridium thermosulfurooenes EM1 which hydrolyses both a-l,6 and a-l,4-glycosidic linkages. Appl. Microbiol. Biotechnol. 1990, 33, 511−518. (19) Ali, G.; Rihouey, C.; Le Cerf, D.; Picton, L. Effect of carboxymethyl groups on degradation of modified pullulan by pullulanase from Klebsiella pneumoniae. Carbohydr. Polym. 2013, 93, 109−115. (20) Belbekhouche, S.; Ali, G.; Dulong, V.; Picton, L.; Le Cerf, D. Synthesis and characterization of thermosensitive and pH-sensitive block copolymers based on polyetheramine and pullulan with different length. Carbohydr. Polym. 2011, 86, 304−312. (21) Tan, K. Y. D.; Kee, J. W.; Fan, W. Y. CpMn(CO)3-Catalyzed Photoconversion of Thiols into Disulfides and Dihydrogen. Organometallics 2010, 29, 4459−4463. (22) Thebault, P.; Taffin de Givenchy, E.; Levy, R.; Vandenberghe, Y.; Guittard, F.; Géribaldi, S. Preparation and antimicrobial behaviour of quaternary ammonium thiol derivatives able to be grafted on metal surfaces. Eur. J. Med. Chem. 2009, 44, 717−724. (23) Cleland, W. W. Dithiothreitol, a New Protective Reagent for SH Groups*. Biochemistry 1964, 3, 480−482. (24) Simons, S. S., Jr.; Johnson, D. F. Reaction of o-Phthalaldehyde and Thiols with Primary Amines: Formation of 1-Alkyl (and aryl)thio2-alkylisoindole. J. Org. Chem. 1978, 43, 2886−2891. (25) Freedman, R. B.; Radda, G. K. The Reaction of 2,4,6Trinitrobenzenesulphonic Acid with Amino Acids, Peptides and Proteins. Biochem. J. 1968, 108, 383−391. (26) Clayton, J. W.; Meredith, W. O. S. The effect of thiols on the dinitrosalicylic acid test for reducing sugars. J. Inst. Brew. 1966, 72, 537−540. (27) Hill, H. D.; Straka, J. G. Protein Determination Using Bicinchoninic Acid in the Presence of Sulfhydryl Reagents. Anal. Biochem. 1988, 170, 203−208. (28) Bain, C. D.; Troughton, E. B.; Tao, Y.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. Formation of Monolayer Films by the Spontaneous Assembly of Organic Thiols from Solution onto Gold. J. Am. Chem. Soc. 1989, 111, 821−335. (29) Humblot, V.; Yala, J. F.; Thebault, P.; Bourkema, K.; Héquet, A.; Berjeaud, J. M.; Pradier, C. M. The antibacterial activity of Magainin I immobilized onto mixed thiols Self-Assembled Monolayers. Biomaterials 2009, 30, 3503−3512. (30) Wang, S.; van Dijk, J. A. P. P.; Odijk, T.; Smit, J. A. M. Depletion-Induced Demixing in Aqueous Protein-Polysaccharide Solutions. Biomacromolecules 2001, 2, 1080−1088. (31) Cao, X.; Pettit, M. E.; Conlan, S. L.; Wagner, W.; Ho, A. D.; Clare, A. S.; Callow, J. A.; Callow, M. E.; Grunze, M.; Rosenhahn, A. Resistance of Polysaccharide Coatings to Proteins, Hematopoietic Cells, and Marine Organisms. Biomacromolecules 2009, 10, 907−915. (32) Pertile, R. A. N.; Andradea, F. K.; Alves, C., Jr.; Gamaa, M. Surface modification of bacterial cellulose by nitrogen-containing plasma for improved interaction with cells. Carbohydr. Polym. 2010, 82, 692−698.

(33) Wang, Y. W.; Fan, L. J. High-Resolution XPS Study of Decanethiol on Au(111): Single Sulfur-Gold Bonding Interaction. Langmuir 2002, 18, 1157−1164. (34) Ishida, T.; Choi, N. High-Resolution X-ray Photoelectron Spectra of Organosulfur Monolayers on Au(111): S(2p) Spectral Dependence on Molecular Species. Langmuir 1999, 15, 6799−6806. (35) Zhong, C. J.; Brush, R. C.; Anderegg, J.; Porter, M. D. Organosulfur Monolayers at Gold Surfaces: Reexamination of the Case for Sulfide Adsorption and Implications to the Formation of Monolayers from Thiols and Disulfides. Langmuir 1999, 15, 518−525. (36) Wirde, M.; Gelius, U. Self-Assembled Monolayers of Cystamine and Cysteamine on Gold Studied by XPS and Voltammetry. Langmuir 1999, 15, 6370−6378. (37) Telmer, P. G.; Shilton, B. H. Insights into the conformational equilibria of maltose-binding protein by analysis of high affinity mutants. J. Biol. Chem. 2003, 278, 34555−34567.

261

DOI: 10.1021/la504212k Langmuir 2015, 31, 254−261