Improving the Surface Biocompatibility with the ... - ACS Publications

ARTICLE pubs.acs.org/Langmuir

Improving the Surface Biocompatibility with the Use of Mixed Zwitterionic Self-Assembled Monolayers Prepared by a Proper Solvent Ching-Hsiung Shen and Jui-Che Lin* Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan 70101

bS Supporting Information ABSTRACT: In this study, the mixed self-assembled monolayers (SAMs) containing the mixture of long-chain alkanethiol, SH(CH2)11NH2 and SH(CH2)10SO3H, was prepared as a model surface to examine the interaction between the biological environment and artificial surface. The 10% (v/v) NH4OH ethanolic solution and DMSO were chosen as the solvents for the preparation of these mixed SAMs and the “solvent effect” was discussed. X-ray photoelectron spectroscopy (XPS) has indicated that
SO3H/
NH2 mixed SAMs formed from 10% (v/v) NH4OH ethanolic solution were surface “
SO3H poor”, while a nearly equivalent amount of surface
SO3H functionality was presented on the mixed SAMs formed from DMSO. This has resulted from the different solvation capability between solvent molecules and the alkanethiol. Such solvent effects were also reflected in various surface properties such as surface wettability and surface zeta potential. The mixed SAMs formed from DMSO were more surface hydrophilic and less negatively surface charged than from 10% (v/v) NH4OH ethanolic solution. In addition, these mixed SAMs formed from DMSO exhibited the least amount of protein adsorbed as well as a better platelet compatibility than its counterpart from 10% (v/v) NH4OH ethanolic solution. These findings indicated that choosing a proper solvent for mixed zwitterionic SAM can greatly affect its surface properties and biocompatibility, such as to form a surface with near neutrality for reducing protein adsorption and subsequent platelet adhesion and activation.

1. INTRODUCTION Blood-contacting biomedical devices such as artificial heart vessels, heart valves, and stents may suffer from the problems of thrombus formation, caused by the series of protein adsorption, platelet adhesion, and activation.1,2 The interfacial interactions between the biological environment and the biomaterial were controlled by the surface properties of the synthetic material.3,4 Hence, optimization of the surface properties of the artificial biomaterial has become the major area of research in improving its blood compatibility. Recently, self-assembled monolayers (SAM) prepared by long-chain alkanethiols (HS-(CH2)nX, n g 10) on gold can offer a densely packed, well-defined, and highly ordered surface with variant functional groups.5,6 Because of these prominent advantages, this can serve as a model surface to study the blood
material interaction.2,7
10 Several investigators have shown that polymer surfaces containing sulfonate or sufonic acid functionality can enhance the blood compatibility and it may be attributed to “heparin like structure”, containing sulfonic acid functionality along the polymer backbone.11,12 In our previous study, we presented that platelet reactivity of sulfonic acid terminated SAM was higher than that for the
OH terminated one. It might be caused by the higher densities of surface functionalities, the intrinsic properties r 2011 American Chemical Society

associated with the SAMs prepared by single alkanethiol.2 With the addition of the technology of mixed SAMs, the surface densities can be regulated by mixing different alkanethiols in the preparation solvent.13
15 Two different kinds of mixed SAMs that contained the
SO3H (hydrophilic, negative charge) terminated alkanethiol and the
CH3 (hydrophobic, neutral charge) or
OH (hydrophilic, neutral charge) terminated one were made. Although they can successfully adjust the surface density of
SO3H terminated SAMs, these mixed SAMs cannot effectively improve the blood compatibility of the
SO3H terminated one. This was caused by the fairly negative values of surface zeta potential that induce protein adsorption and subsequent platelet adhesion and activation.10,16 In this study, the
NH2 terminated thiol (positive charge) was added to form the
SO3H and
NH2 mixed SAMs with an attempt to balance the negative charges on the surface. The charge-balanced surfaces with equal valence of positively and negatively charged functionality had been implicated in promoting better blood compatibility.8,17,18 Received: March 10, 2011 Revised: May 3, 2011 Published: May 12, 2011 7091

dx.doi.org/10.1021/la200906b | Langmuir 2011, 27, 7091–7098

Langmuir Because the nature of the solvent may influence the kinetics of formation and the mechanisms of self-assembly for preparing the SAM of alkanethiol, the packing structure and interfacial properties of the resultant SAM would be varied by the choice of preparation solvent.6 Bain et al. have proposed that when preparing the mixed SAMs the interactions among the functional groups as well as the interactions between the functional groups and solvent will play a crucial role to affect its surface composition.13,19 Mostly ethanol, a polar protic solvent, was used to prepare the SAM. In contrast, dimethyl sulfoxide (DMSO), a polar nonprotic solvent with good solubility to
NH2 and
SO3H terminated alkanethiol, was rarely chosen as the preparation solvent. When preparing the SAM with cationic amine functionality, the thiols must assemble from the basic ethanolic solution.8,20,21 It was because the addition of little base triethylamine or ammonium hydroxide can disrupt the hydrogen bond and suppress the ionization of the NH2 group to form the better packing quality of SAM. Then, the washing step with the additional usage of 10% CH3COOH or 1% HCl ethanolic solution to remove the adsorbed free thiol and base triethylamine or ammonium hydroxide can lead to a monolayer of
NH2 SAM with good quality. In view of these likely solvent effects, 10% NH4OH ethanolic solution and DMSO were chosen to prepare the
SO3H and
NH2 mixed SAMs. Then, 1% HCl ethanolic solution was utilized for the washing step. The “solvent effect” will be examined to find the appropriate solvent for preparing the mixed zwitterionic SAMs with suitable surface characteristics. The surface characteristics of these SAMs will be determined by contact angle measurement, X-ray photoelectron spectroscopy (XPS), and streaming potential measurement within the PBS at pH 7.4. Furthermore, surface plasmon resonance (SPR) was used to clarify the protein adsorption and in vitro platelet adhesion assay using the platelet-rich plasma as the incubation medium was also evaluated. Finally, the correlation among the surface properties, protein adsorption results, and platelet compatibility will be discussed.

2. MATERIALS AND METHODS 2.1. Synthesis of 10-Mercaptodecanesulfonic Acid and (11-Mercaptodecyl) Ammonium Chloride. All reagents and solvents used were either HPLC or reagent grade, and were used without further purification. The synthesis strategy was adopted from our previously published procedures.2,8 2.2. Preparation of Gold Substrates. The substrates used in this experiment were Si wafers and cover glass. The Si wafers were used in contact angle measurements, X-ray photoelectron spectroscopy (XPS), and in vitro platelet adhesion assay. The cover glass was used in zeta potential measurements. The Si wafers and cover glasses were cut into 3 cm  1.5 cm size and 5.5 cm  2.5 cm size, respectively. They were first cleaned by piranha solution (3:7 vol ratio of 30 wt % hydrogen peroxide solution and concentrated sulfuric acid) at 90 °C for 2 h. (Caution: The piranha solution was violently corrosive to organic matter and should be handled with great care). Then, they were thoroughly rinsed with a large amount of deionized water and absolute ethanol and then blown dry with nitrogen. These cleaned substrates were immediately put into vacuum chamber of thermal evaporator for physical vapor deposition of Ti (Au adhesion promotion layer) and Au. The thicknesses of Ti and Au were 150 Å and 1000 Å, respectively.

2.3. Preparation of Mixed Zwitterionic Self-Assembled Monolayers. The freshly prepared gold substrates were immediately

ARTICLE

immersed into two different mixed alkanethiol solutions, containing the mixture of the 10-mercaptodecanesulfonate and 11-mercaptodecyl ammonium chloride in DMSO or absolute ethanol containing 10% (v/v) NH4OH at room temperature for 24 h. The mole fraction of
SO3H terminated alkanethiol in mixed alkanethiol solutions was 0, 0.3, 0.5, 0.7, and 1. The total concentration of the solution was set at 1 mM. For the mixed SAMs prepared by absolute ethanol containing 10% (v/v) NH4OH, these substrates were sequentially rinsed with absolute ethanol, absolute ethanol containing 1% HCl, and absolute ethanol. For the mixed SAMs prepared by DMSO, these substrates were sequentially rinsed with absolute ethanol, deionized water, absolute ethanol containing 1% HCl, and absolute ethanol. Then, they were blown dry with nitrogen. 2.4. Surface Characterization. The surface wettability was characterized by sessile drop technique (Face, model CA-A; Tokyo, Japan) at room temperature. The probing liquid was deionized water. The measurement was carried out at four different spots for each substrate and each spot was repeated 3 times. The surface composition measurement was carried out by X-ray photoelectron spectroscopy (XPS or ESCA) (PHI Quantera SXM, USA). The X-ray source was monochromatic Al KR (hν = 1486.6 eV) and the takeoff angle was set at 45°. The high-resolution spectra were deconvoluted by mixing Gaussian
Lorentzian functions using the free software XPSPEAK. Quantification of the element was done on the peak areas of the C 1s, O 1s, N 1s, and S 2p multiplex. The zeta potential of flat surfaces was determined by SurPass ElectroKinetic analyzer (Anton Paar KG, Graz, Austria). To mimic the physiological conditions, phosphate buffer saline (PBS) at pH 7.4 was used as the electrolyte solution. When the electrolyte solution was flowed over the flat substrates under moderate pressure, the streaming potential was measured by the two Ag/AgCl electrodes, placed at the inlet and outlet of the fluid cell. Zeta potential ζ was calculated from streaming potential, ΔUs, using the Helmholtz
Smoluchowski equation and Fairbrother
Mastin approach:22,23 ζ¼

ΔUs kη ΔP εε0

where (ΔUs)/(ΔP) is the slope of the curve between the streaming potential vs the increasing applied potential through the streaming cell, η and ε are the viscosity and permittivity of the streaming electrolyte, ε0 is the permittivity of the free space, and k is the solution conductivity. 2.5. Protein Adsorption Experiments. The custom-built SPR biosensor based on wavelength interrogation with four-channel Teflon flow cell was used to measure the protein adsorption on these mixed SAMs, and the experimental protocol was similar to that used in previous publications.24
26 First, bovine serum albumin (BSA, Sigma-Aldrich, USA) and bovine fibrinogen (Calbiochem, USA) were dissolved in phosphate-buffered saline (PBS, 0.15 M, pH 7.4, Sigma-Aldrich, USA), and the final albumin and fibrinogen concentration was 1 mg/mL and 0.1 mg/mL, respectively. The mixed protein solution was prepared by mixing 2 mg/mL BSA and 0.2 mg/mL bovine fibrinogen in equal volume. Before the protein adsorption, the chips were flowed with PBS until the system was in equilibrium. In this work, three different protein adsorption experiments were made: (1) 1 mg/mL BSA was adsorbed for 60 min; (2) 0.1 mg/mL bovine fibrinogen adsorbed for 60 min; (3) mixed protein solution (1 mg/mL BSA and 0.1 mg/mL bovine fibrinogen) adsorbed for 60 min. After the above-mentioned protein adsorption for 60 min, the chip was washed with PBS for 30 min. The measured wavelength shift subtracted by the wavelength shift of the baseline was the real wavelength shift caused by the protein adsorption. 2.6. In Vitro Platelet Adhesion Assay. The procedures were followed from our previous studies.8
10 Briefly, as shown below, the substrates were pre-equilibrated in Petri dishes containing HepesTyrodes solution for 2 h. The human platelet-rich plasma was carefully 7092

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098

Langmuir

ARTICLE

ethanolic solution except when XSO3H,soln = 0 and 1. The contact angle of pure
NH2 SAM was similar between these two different solvent-prepared samples, while pure
SO3H SAM formed from 10% NH4OH (v/v) ethanolic solution was more hydrophilic than that formed from DMSO. 3.2. XPS. Table 1 showed the surface atomic composition, as calculated by XPS, of the
SO3H/
NH2 mixed SAMs. The surface atomic percentage of C 1s and N 1s for the SAMs prepared in 10% NH4OH (v/v) ethanolic solution were decreased with the XSO3H,soln, while the O 1s and S 2p were increased. However, for the mixed SAMs prepared by DMSO, the surface atomic percentages of C 1s, N 1s, O 1s, and S 2p were almost the same except for the two pure
NH2 and
SO3H terminated SAMs. On the pure
NH2 terminated SAM (XSO3H,soln = 0), oxygen was noted on those prepared by the two different solvents studied. It might have resulted from the adsorbates containing oxygen such as water tightly bound with the amine headgroup.27 The result shown in Table 1 also indicated that the surface atomic composition of S 2p was smaller than the theoretical values; it was attributed to the inelastic scattering of the S 2p photoelectrons from the thiolate binding that traveled through the monolayer from near-bottom Au deposited into the detector.28 The S 2p spectra of the pure
NH2 SAM (XSO3H,soln = 0) formed from 10% NH4OH (v/v) ethanolic solution and DMSO can be deconvoluted into two peaks, the bound thiolate (162 eV) and unbound thiol (163.5
164 eV)29 (Figure S1 in the Supporting Information). It meant that these two preparing solvents can form the monolayer for
NH2 terminated SAM successfully. As the
SO3H thiol was added, an additional S 2p peak centered at 168
168.6 eV appeared and it can be assigned to the sulfonic acid functionality.2 The N 1s spectra of these mixed SAMs was also deconvoluted into two peaks, the protonated amine (NH3þ, 401
402 eV) and deprotonated amine (NH2, 399.5
400 eV)30 (Figure S2 in the Supporting Information). As the XSO3H,soln increases, the area of sulfonic acid and protonated amine functionality were increased on the mixed SAMs formed from 10% (v/v) NH4OH ethanolic solution. However, for the mixed SAMs prepared from DMSO the area of sulfonic acid and protonated amine functionality remained nearly the same. The calculation of the surface mole fraction of
SO3H terminated SAM (XSO3H,surf) was based on the equation below:17

added into the Petri dishes and then placed inside the incubator with CO2 flow (5%) at 37 °C for 1 h. Following that, the substrates were gently rinsed with Hepes-Tyrodes solution three times and fixed with Hepes solution containing 2% (v/v) glutaraldehyde for 30 min. Then, these substrates were rinsed sequentially with Hepes-Tyrodes solution of 100%, 75%, 50%, 25%, and deionized water, and then dehydrated with ethanol of 25%, 50%, 75%, and 100% for 3 min each by a shaker at 50 rpm. The substrates were dried by a CO2 critical point dryer (CPD) (Hitachi HCP-2, Japan) and immediately coated with a layer of Au. The densities of the adhered platelets were calculated from three different regions for each substrate at 1000 magnification. The platelet adhesion experiments were repeated three times for each sample.

3. RESULTS 3.1. Contact Angle. The contact angle of
SO3H/
NH2 mixed SAMs as the function of mole fraction of
SO3H thiol (XSO3H,soln) formed from 10% NH4OH (v/v) ethanolic solution and DMSO was shown in Figure 1. As the XSO3H,soln increased, the contact angle value was gradually decreased for the SAMs prepared in 10% NH4OH (v/v) ethanolic solution. The contact angle of the mixed SAMs prepared from DMSO was also decreased with the increase in XSO3H,soln but increased dramatically when XSO3H,soln = 1 (pure
SO3H SAMs), even higher than the pure
NH2 (XSO3H,soln = 0) terminated SAMs. In addition, the contact angle of the mixed SAM formed from DMSO was lower than that formed from 10% NH4OH (v/v)

XSO3H, surf ¼ Figure 1. Contact angle of
SO3H/
NH2 mixed SAMs as the function of mole fraction of
SO3H thiol (XSO3H,soln) prepared by 10% NH4OH (v/v) ethanolic solution and DMSO. Data shown is mean ( SD (n = 3).

ðASO3H ÞXSO3H, soln ðASO3H ÞXSO3H, soln þ ðAN ÞXSO3H, soln

where ASO3H was the normalized area of the sulfonic acid functionality in the S 2p spectra and AN was the normalized area of the nitrogen-containing functionality in the N 1s spectra.

Table 1. Surface Composition of
SO3H/
NH2 Mixed SAMs Prepared by 10% NH4OH (v/v) Ethanolic Solution and DMSOa 10% NH4OH XSO3H,soln

a

DMSO

C (%)

O (%)

S (%)

N (%)

C (%)

O (%)

S (%)

N (%)

0

85.27 (78.57)

5.27 (0.00)

2.66 (7.14)

6.81 (7.14)

0.3

80.69 (75.00)

8.7 (6.00)

4.58 (9.00)

6.03 (5.00)

82.98 (78.57)

7.10 (0.00)

4.08 (7.14)

5.84 (7.14)

76.46 (75.00)

13.39 (6.00)

6.31 (9.00)

0.5

78.60 (72.62)

11.81 (10.00)

4.32 (10.24)

5.27 (3.57)

3.84 (5.00)

74.84 (72.62)

13.83 (10.00)

6.12 (10.24)

5.21 (3.57)

0.7

77.82 (70.24)

11.60 (14.00)

5.68 (11.48

4.90 (2.14)

76.80 (70.24)

13.02 (14.00)

5.69 (11.48

4.49 (2.14)

1

70.20 (66.67)

22.66 (20.00)

7.14 (13.33)

0.00 (0.00)

65.22 (66.67)

27.93 (20.00)

6.85 (13.33)

0.00 (0.00)

Values listed within the parentheses are calculated by the stoichiometric ratio of the alkanethiol within the solution. 7093

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098

Langmuir

Figure 2. Relationship between the surface mole fraction of
SO3H thiol (XSO3H,surf) and the solution mole fraction of
SO3H thiol (XSO3H,soln) of
SO3H/
NH2 mixed SAMs prepared by 10% NH4OH (v/v) ethanolic solution and DMSO.

Figure 3. Zeta potential of
SO3H/
NH2 mixed SAMs prepared by 10% NH4OH (v/v) ethanolic solution and DMSO. Data shown is mean ( SD (n = 3).

From the correlation plot of solution mole fraction of
SO3H terminated thiol (XSO3H,soln) and the surface mole fraction of
SO3H terminated SAM (XSO3H,surf) (Figure 2), the SAMs formed from the 10% (v/v) NH4OH ethanolic solution showed “
SO3H alkanethiol poor” behavior. On the contrary, the SAMs formed from DMSO showed a sigmoidal shape which is quite different the preferential adsorption noted on the SAMs formed from the ethanol containing 10% NH4OH. 3.3. Zeta Potential. The zeta potential of these SAMs prepared by 10% (v/v) NH4OH ethanolic solution and DMSO as determined by streaming potential in PBS (pH 7.4, ionic strength = 140 mM) was all negative and the pure
SO3H SAM (XSO3H,soln = 1) has the most negative zeta potential (Figure 3). In comparison to the zeta potential prepared by different solvent, the zeta potential of pure
SO3H and pure
NH2 terminated SAM did not vary with the solvent used. Contrarily, the mixed SAMs formed from DMSO showed less negative surface charge than those formed from 10% (v/v) NH4OH ethanolic solution.

ARTICLE

3.4. In Vitro Protein Adsorption Experiments. The amount of proteins adsorbed on
SO3H/
NH2 mixed SAMs prepared by 10% NH4OH (v/v) ethanolic solution was shown in Figure 4a. The amount of BSA adsorbed was decreased with the increase of XSO3H,soln. However, for the bovine fibrinogen and mixed protein, the trend was different. The ranked order was XSO3H,soln = 0 > XSO3H,soln = 0.3 > XSO3H,soln = 1 > XSO3H,soln = 0.5 > XSO3H,soln = 0.7 for bovine fibrinogen and XSO3H,soln = 0.3 > XSO3H,soln = 1 > XSO3H,soln = 0.5 > XSO3H,soln = 0 > XSO3H,soln = 0.7 for mixed protein, respectively. However, in Figure 4b it was found that the SPR wavelength shifts of BSA, bovine fibrinogen, and mixed protein adsorbed on the mixed SAMs prepared by DMSO were all smaller than on the pure
NH2 and
SO3H terminated SAMs. In addition, the amounts of BSA, bovine fibrinogen, and mixed protein adsorbed on the mixed SAMs prepared in DMSO were all significantly smaller than that of its counterpart prepared in 10% (v/v) ethanolic solution. 3.5. In Vitro Platelet Adhesion Assay. The densities of adherent platelets on
SO3H/
NH2 mixed SAMs prepared by different solvents were shown in Figure 5. On the SAMs formed from 10% NH4OH (v/v) ethanolic solution, pure
SO3H SAM exhibited the highest platelet adhesion density while others did not vary significantly. However, the densities of adherent platelets on the mixed SAMs formed from DMSO were all lower than on the pure
SO3H and
NH2 SAMs. Moreover, the platelet adhesion densities on the pure
NH2 and mixed SAMs formed from DMSO were all dramatically lower than that of its counterpart from 10% NH4OH ethanolic solution. The morphologies of adherent platelets on these SAMs were all less activated (i.e., round and dendritic) than on the Au control (Figure S3 in the Supporting Information).

4. DISCUSSION Our previous study had shown that the pure
SO3H terminated SAM displayed more adherent platelets due to the high surface functional group density.2 Therefore, the
CH3 terminated thiol (hydrophobic, neutral charge) and
OH terminated one (hydrophilic, neutral charge) were added to regulate the surface density of the sulfonic acid functionality.10 However, the mixed SAMs containing the
SO3H terminated thiol with
CH3 or
OH terminated one still exhibited more adherent platelets except on the pure
OH terminated one. It resulted from the more negative surface charge to induce more protein adsorption and further caused more platelet adhesion. In this study, the zwitterionic mixed SAMs were prepared by adding the
NH2 terminated thiol (positive charge) into the
SO3H SAM (negative charge) to manipulate the density of
SO3H functionality and the surface charge. The solvent played a predominant role in the formation of the mixed SAMs and could influence the surface composition of the specific thiol.13,19 Mostly, the chosen solvent in which thiol self-assembled was ethanol, yet in this study, the DMSO was also used. In Figure 2, the relationship between XSO3H,soln and XSO3H,surf was “
SO3H alkanethiol poor” on the
SO3H/
NH2 mixed SAMs formed from 10% NH4OH (v/v) ethanolic solution. However, on the mixed SAMs formed from DMSO it was shown that
NH2 thiol still preferentially adsorbed even as the XSO3H,soln was more than 0.5. It was the so-called ideal nonideality behavior in that the surface composition was independent of the solution composition and was similar to previous studies.17,30 The difference of these two diverse behaviors was caused by the 7094

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098

Langmuir

ARTICLE

Figure 4. Amounts of BSA, bovine fibrinogen, and mixed protein adsorbed on
SO3H/
NH2 mixed SAMs prepared by (a) 10% NH4OH (v/v) ethanolic solution and (b) DMSO, respectively.

Figure 5. Densities of adherent platelets on
SO3H/
NH2 mixed SAMs as the function of XSO3H,soln prepared by 10% NH4OH (v/v) ethanolic solution and DMSO. Data shown is mean ( SD (n = 3).

solvation capability between alkanethiols and solvent molecules. It implicates a similar solubility for the
SO3H terminated thiol and
NH2 terminated one in DMSO but different solubility of two thiols in ethanol. Moreover, on the mixed SAMs formed from 10% (v/v) NH4OH ethanolic solution the area percentage of protonated amine (NH3þ, 401
402 eV) increased with area percentage of sulfonic acid. (Figure S1 (a) and S2 (a) in Supporting Information). It was likely resulted from the electrostatic interaction between
SO3
and
NH3þ as the XSO3H,surf increases Contrarily, the area percentages of sulfonic acid and protonated amine of the mixed SAMs formed from DMSO were almost similar among each other (Figure S1 (b) and S2 (b) in Supporting Information). This was because the XSO3H,surf was independent of the increase of XSO3H,soln (Figure 2). Therefore, these results further proved that the self-assembly process could be affected by the differences in solvation capability for
NH2 and
SO3H terminated alkanethiol. Such differences among the terminal functionalities and solvent molecules may lead to variations in surface adsorption and thiolate formation rate, and eventually affect the surface composition.

The surface hydrophilicities of the SAMs formed from 10% (v/v) NH4OH ethanolic solution and DMSO exhibited different patterns (Figure 1). On XSO3H,soln = 0 (pure
NH2 SAM) prepared in two different solvents, the contact angle values were similar. In contrast, the contact angle of XSO3H,soln = 1 (pure
SO3H SAM) formed from 10% (v/v) NH4OH ethanolic solution was lower than that formed from DMSO, which might be caused by the variations in packing structure affected by different solvent molecules. However, the mixed SAMs formed from DMSO were more hydrophilic than of those formed from 10% NH4OH (v/v) ethanolic solution. The difference could be from the different surface composition and/or nanoscale variations during the self-assembly process in different solvent. On
NH2/
SO3H mixed SAMs formed from 10% NH4OH (v/v) ethanolic solution and DMSO, the zeta potentials in PBS at pH 7.4 were all negative (Figure 3). Because the sulfonic acid functionality is a strongly dissociated acid group, the zeta potential was negative10 and did not show significant variation with the type of solvent used for preparing the pure
SO3H SAM (XSO3H,soln = 1). The literature has shown that the isoelectric points (IEP) of
NH2 SAM prepared by ethanol and ethanolic solution added with aqueous HCl solution were pH 6.5 and pH 7.2 in 0.1 mM NaCl solution, respectively.31,32 Therefore, the zeta potential for XSO3H,soln = 0 (pure
NH2 SAM) at pH 7.4 in PBS should be negative. In comparison to the zeta potential of mixed SAMs prepared in different solvent, the zeta potential formed from DMSO was less negative than that of its counterpart formed from 10% NH4OH (v/v) ethanolic solution. Previous reports have shown that the zeta potential, the charge at the border between the Stern layer and the diffuse layer, was influenced by the surface chemical composition, surface polarity, and swelling behaviors.33 Therefore, it might be caused by the variation of surface composition of XSO3H,surf in SAM prepared by different solvent. However, for the two mixed SAMs with XSO3H,soln = 0.7, the XSO3H,surf was similar while the zeta potential of XSO3H,soln = 0.7 formed from DMSO was less negative than of that formed from 10% NH4OH (v/v) ethanolic solution. Because the zeta potential was the surface charge at the slipping plane, this might be due to the different ordering/orientation of ions or counterions in the solution caused by the different spatial distribution of functional groups. 7095

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098

Langmuir

ARTICLE

Figure 6. Relationship of protein adsorption with zeta potential and the surface composition of
SO3H terminated alkanethiol (XSO3H,surf) on
SO3H/
NH2 mixed SAMs prepared by (a) 10% NH4OH (v/v) ethanolic solution and (b) DMSO. The SPR wavelength shift, which means the amount of BSA, bovine fibrinogen, and mixed protein adsorption, was presented as a straight strip chart with the left y-axis. The negative zeta potential of mixed SAMs was presented as a line graph with the right y-axis. The inverted triangle symbol showed the surface composition of
SO3H SAM and the values were indicated in the parentheses.

Figure 7. Relationship of protein adsorption with platelet adhesion density on
SO3H/
NH2 mixed SAMs prepared by (a) 10% NH4OH (v/v) ethanolic solution and (b) DMSO. The SPR wavelength shift, which means the amount of BSA, bovine fibrinogen, and mixed protein adsorption, was presented as a straight strip chart with the left y-axis. Platelet adherent density of mixed SAMs was presented as a line graph with the right y-axis.

The behavior of protein adsorption was highly related to the surface physiochemical properties of biomaterials.34
37 From the above-mentioned results, the nature of solvent can affect the formation process of mixed SAMs and, henceforth, change its surface properties. Figure 6 showed the relationship of protein adsorption with zeta potential and XSO3H,surf on these SAMs. For
SO3H/
NH2 mixed SAMs formed from 10% (v/v) ethanolic solution, the amount of BSA decreased with the increase of XSO3H,soln. This was likely caused by the repulsive electrostatic interaction because the BSA and the surface charge of
SO3H functionality were all negative at pH 7.4.35 Nevertheless, the results of three protein experiments were not all directly correlated to the zeta potential. For
SO3H/
NH2 mixed SAMs formed from DMSO, the least amounts of BSA, bovine fibrinogen, and mixed protein adsorbed were noted on XSO3H,soln = 0.3, 0.5, and 0.7, in which similar surface composition of
SO3H functionality (XSO3H,surf = 0.4) was noted, and this value is nearly equal to the surface percentage of
SO3H and
NH2 (i.e., XSO3H,surf = 0.5). Moreover, these three surfaces were also the

least negatively charged surfaces. This indicated that the surface with less negative surface zeta potential would adsorb a lower amount of protein. Previous studies have presented that the surface with a similar amount of cationic and anionic functionality can resist protein adsorption and this was presumably caused by the charge balance effect on protein adsorption.8,17,18 Therefore, the mixed SAMs formed from DMSO can effectively reduce the protein adsorption as compared to those from 10% NH4OH (v/v) ethanolic solution. Henceforth, the nature of solvent will determine the surface composition and surface charge and, subsequently, protein adsorption. In correlating the platelet adhesion density with the adsorbed protein amount, it was noted that the amount of BSA, bovine fibrinogen, and mixed protein adsorbed on the mixed SAMs formed from 10% NH4OH (v/v) ethanolic solution (Figure 7a) was inversely proportional to the platelet adhesion density. However, in Figure 7b it was noted the mixed SAMs formed from DMSO that presented a small amount of protein adsorbed also showed the least platelets adhered. Studies have indicated 7096

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098

Langmuir that the adsorbed fibrinogen can bind with the glycoprotein IIBIIIA (GP IIB-IIIA), the specific fibrinogen receptor on the platelet cell membrane, and results in more platelet adhesion and aggregation.38,39 Therefore, the surface with less fibrinogen adsorption should cause less platelet adhesion as noted on XSO3H,soln = 0.3, 0.5, and 0.7 formed from DMSO. However, it was interesting that for XSO3H,soln = 0.3, 0.5, and 0.7 formed from 10% NH4OH (v/v) ethanolic solution the amount of bovine fibrinogen and mixed protein adsorbed was inversely proportional to the adhered platelet density. The composition and conformation of adsorbed proteins may influence subsequent platelet adhesion and activation.40,41 Therefore, it could be a result from the different conformation of adsorbed proteins that led to the platelet adhesion in spite of fewer bovine fibrinogens and mixed protein adhered. Moreover, the amount of adsorbed protein and adhered platelet density on XSO3H,surf = 0.7 formed from DMSO was lower than from 10% NH4OH in spite of similar XSO3H,surf value. This may be related to differences in the nanoscale surface distribution of cationic and anionic functionality, which were affected by the chosen solvent. In summary, when DMSO was chosen as the solvent, adding the positive
NH2 terminated thiol to the negative
SO3H terminated SAM can effectively regulate the amount of surface charge for preparing surface with less negative charge. This approach can further reduce the bovine fibrinogen and mixed protein adsorption and, consequently, the platelet adhesion and activation.

5. CONCLUSION In this study, the mixed zwitterionic SAM composed of the mixture of negative
SO3H terminated alkanethiol and positive
NH2 terminated one was successfully prepared. The 10% (v/v) NH4OH ethanolic solution and DMSO were chosen as the preparation solvent to characterize the likely “solvent effect” on this mixed SAM. Owing to the variations in the interaction between the solvent molecules and the terminal functional groups, different self-assembling mechanisms as well as distinct surface properties were observed. For the mixed SAMs formed from 10% NH4OH (v/v) ethanolic solution, the “
SO3Hterminal ends-poor” were noted, but for those formed from DMSO, XSO3H,surf was independent of XSO3H,soln used and nearly about 0.4. Furthermore, it was observed that higher surface hydrophilicity and less negative zeta potential at pH 7.4 in PBS for the mixed SAMs formed from DMSO. These findings can be attributed to the different solvation capability between the alkanethiols and the solvent molecules used. The results of protein adsorption experiments have shown that fewer amounts of BSA, bovine fibrinogen, and mixed proteins were adsorbed onto the mixed SAMs formed from DMSO than from 10% NH4OH (v/v) ethanolic solution. This finding was highly correlated to the in vitro platelet adherent densities. Therefore, the surfaces with less negative surface potential and nearly equal amounts of positively and negatively charged functional groups can resist the protein adsorption and, further, platelet adhesion and activation. Henceforth, the nature of solvent played an important role in preparing the mixed zwitterionic SAMs. Adding the positive
NH2 terminated thiol to the negative
SO3H terminated SAMs in DMSO can successfully modulate the composition and spatial distribution of the negative
SO3H functionality to prepare surface with a better blood compatibility.

ARTICLE

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional figures as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Address correspondence to: Jui-Che Lin, Ph.D., Department of Chemical Engineering, National Cheng Kung University, Tainan, TAIWAN 70101. Phone: þ886-6-275-7575 ext. 62665. Fax: þ886-6-234-4496. E-mail: [email protected].

’ ACKNOWLEDGMENT The financial support from the National Science Council, Taiwan under Grants NSC 95-2221-E-006-305, NSC 97-2221E-006-017-MY3, and EZ-12-21-07-99 is gratefully appreciated. We thank the Tainan Blood Donation Center for providing human platelet-rich plasma. We also sincerely thank the R&D Center for Membrane Technology and Prof. Yung Chang at Chung Yuan Christian University for providing the surface plasmon resonance instrument to measure the protein adsorption. ’ REFERENCES (1) Zhang, Z.; Zhang, M.; Chen, S. F.; Horbetta, T. A.; Ratner, B. D.; Jiang, S. Y. Biomaterials 2008, 29 (32), 4285–4291. (2) Lin, J. C.; Chuang, W. H. J. Biomed. Mater. Res. 2000, 51 (3), 413–423. (3) Elbert, D. L.; Hubbell, J. A. Annu. Rev. Mater. Sci. 1996, 26, 365–394. (4) Roach, P.; Eglin, D.; Rohde, K.; Perry, C. C. J. Mater. Sci.: Mater. Med. 2007, 18 (7), 1263–1277. (5) Ulman, A. Chem. Rev. 1996, 96 (4), 1533–1554. (6) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105 (4), 1103–1169. (7) Tsai, M. Y.; Lin, J. C. J. Biomed. Mater. Res. 2001, 55 (4), 554–565. (8) Chuang, W. H.; Lin, J. C. J. Biomed. Mater. Res., A 2007, 82A (4), 820–830. (9) Tsai, M. Y.; Sun, Y. T.; Lin, J. C. J. Colloid Interface Sci. 2007, 308 (2), 474–484. (10) Shen, C. H.; Lin, J. C. Colloids Surf., B 2010, 79 (1), 156–163. (11) Grasel, T. G.; Cooper, S. L. J. Biomed. Mater. Res. 1989, 23 (3), 311–338. (12) Okkema, A. Z.; Visser, S. A.; Cooper, S. L. J. Biomed. Mater. Res. 1991, 25 (11), 1371–1395. (13) Bain, C. D.; Evall, J.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111 (18), 7155–7164. (14) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111 (18), 7164–7175. (15) Allara, D. L. Biosens. Bioelectron. 1995, 10 (9
10), 771–783. (16) Shen, C. H.; Chen, Z. Y.; Chang, Y.; Lin, J. C. Thin Solid Films. (17) Chen, S. F.; Yu, F. C.; Yu, Q. M.; He, Y.; Jiang, S. Y. Langmuir 2006, 22 (19), 8186–8191. (18) Holmlin, R. E.; Chen, X. X.; Chapman, R. G.; Takayama, S.; Whitesides, G. M. Langmuir 2001, 17 (9), 2841–2850. (19) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110 (19), 6560–6561. (20) Li, L. Y.; Chen, S. F.; Jiang, S. Y. Langmuir 2003, 19 (8), 3266–3271. (21) Wang, H.; Chen, S. F.; Li, L. Y.; Jiang, S. Y. Langmuir 2005, 21 (7), 2633–2636. 7097

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098

Langmuir

ARTICLE

(22) Childress, A. E.; Elimelech, M. J. Membr. Sci. 1996, 119 (2), 253–268. (23) Kershner, R. J.; Bullard, J. W.; Cima, M. J. Langmuir 2004, 20 (10), 4101–4108. (24) Chang, Y.; Shu, S. H.; Shih, Y. J.; Chu, C. W.; Ruaan, R. C.; Chen, W. Y. Langmuir 2010, 26 (5), 3522–3530. (25) Chang, Y.; Liao, S. C.; Higuchi, A.; Ruaan, R. C.; Chu, C. W.; Chen, W. Y. Langmuir 2008, 24 (10), 5453–5458. (26) Jung, L. S.; Campbell, C. T.; Chinowsky, T. M.; Mar, M. N.; Yee, S. S. Langmuir 1998, 14 (19), 5636–5648. (27) Baio, J. E.; Weidner, T.; Brison, J.; Graham, D. J.; Gamble, L. J.; Castner, D. G. J. Electron. Spectrosc. 2009, 172 (1
3), 2–8. (28) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111 (1), 321–335. (29) Castner, D. G.; Hinds, K.; Grainger, D. W. Langmuir 1996, 12 (21), 5083–5086. (30) Ooi, Y.; Hobara, D.; Yamamoto, M.; Kakiuchi, T. Langmuir 2005, 21 (24), 11185–11189. (31) Lin, W. C.; Lee, S. H.; Karakachian, M.; Yu, B. Y.; Chen, Y. Y.; Lin, Y. C.; Kuo, C. H.; Shyue, J. J. Phys. Chem. Chem. Phys. 2009, 11 (29), 6199–6204. (32) Lee, S. H.; Lin, W. C.; Kuo, C. H.; Karakachian, M.; Lin, Y. C.; Yu, B. Y.; Shyue, J. J. J. Phys. Chem. C 2010, 114 (23), 10512–10519. (33) Bellmann, C.; Caspari, A.; Loan Doan, T. T.; M€ader, E.; Laxbacer, T.; Kohl, R. Electrokinetic Properties of Natural Fibres, Technical Report; Anton Parr GmbH, Graz, Austria. (34) Sadana, A. Chem. Rev. 1992, 92 (8), 1799–1818. (35) Sigal, G. B.; Mrksich, M.; Whitesides, G. M. J. Am. Chem. Soc. 1998, 120 (14), 3464–3473. (36) Ma, C. F.; Hou, Y.; Liu, S.; Zhang, G. Z. Langmuir 2009, 25 (16), 9467–9472. (37) Nakanishi, K.; Sakiyama, T.; Imamura, K. J. Biosci. Bioeng. 2001, 91 (3), 233–244. (38) Tsai, W. B.; Grunkemeier, J. M.; McFarland, C. D.; Horbett, T. A. J. Biomed. Mater. Res. 2002, 60 (3), 348–359. (39) George, J. N.; Nurden, A. T.; Phillips, D. R. Platelet membrane glycoproteins; Plenum Press: New York, 1985; p xix. (40) Courtney, J. M.; Lamba, N. M. K.; Sundaram, S.; Forbes, C. D. Biomaterials 1994, 15 (10), 737–744. (41) Tsai, W. B.; Grunkemeier, J. M.; Horbett, T. A. J. Biomed. Mater. Res. 1999, 44 (2), 130–139.

7098

dx.doi.org/10.1021/la200906b |Langmuir 2011, 27, 7091–7098