Interactions at the Silica–Peptide Interface - American Chemical Society

Dec 13, 2013 - Interdisciplinary Biomedical Research Centre, School of Science and Technology, Nottingham Trent University, Clifton Lane,. Nottingham ...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/Langmuir

Interactions at the Silica−Peptide Interface: The Influence of Particle Size and Surface Functionality Valeria Puddu and Carole C. Perry* Interdisciplinary Biomedical Research Centre, School of Science and Technology, Nottingham Trent University, Clifton Lane, Nottingham NG11 8NS, United Kingdom S Supporting Information *

ABSTRACT: The variety of interactions that can occur at the silica/aqueous interface makes silica nanoparticles (SiNPs) attractive materials for technological applications. Despite their importance, interfacial interactions are not fully understood. In this contribution, we investigate the effect of (1) particle size and (2) surface functionalization on the adsorption of small biomolecular binders on SiNPs. Small silica binding peptides with different properties (charge, pI, and amino acid composition) were used as binders, while a range of fully characterized SiNPs of diameters ranging between 28 and 500 nm (pristine silica) and SiNPs of ca. 500 nm functionalized with cationic 3-aminopropyl groups and hydrophobic methyl groups was used as binding substrates. Adsorption and binding affinity were investigated by a fluorimetric assay at pH 7.4. A detailed characterization of the surface chemistry of the particles showed that the extent of surface functionalization on modified silica was well below monolayer coverage [by X-ray photoelectron spectroscopy (XPS), ca. 2 and 18 atomic % for the amino- and methyl-modified surfaces, respectively]. Although peptide binding is generally moderated by the physicochemical characteristics of the adsorbing peptide, the introduction of such a small degree of functionality onto silica particles was sufficient to produce drastic changes in adsorption at the silica/aqueous interface. In addition, an increase in peptide adsorption with an increasing particle size, independent of the nature and properties of the peptide, was observed. This particle size effect is attributed to a shift of the dominant binding mechanism toward electrostatic interactions on larger SiNPs. The data presented demonstrate that particle size and surface functionality are both parameters that can substantially influence (bio)molecule uptake via modulation or selection of specific binding modes at the silica/peptide interface. These results are applicable to the design and development of interfaces with specific adsorption/affinity response for biomedical applications, where uptake is important, such as drug delivery. Further, they provide important insights on how peptide affinity and selection during biopanning can be determined by small changes in surface chemistry of the surface of a target that can, in some instances, be associated with the presence of impurities.

1. INTRODUCTION Events occurring at the solid/aqueous interface (i.e., molecular recognition, adsorption, desorption, etc.) underpin a variety of technologies used in the biomedical and biotechnological fields. The use of nanoparticles and multifunctional nanoparticles that combine recognition and targeting with specific properties is widespread for the development of clinical diagnostic tools or therapeutic platforms.1,2 Molecular coating with antibodies1 or small (bio)molecules, such as peptides,3 can be applied on the surface of nanoparticles to improve biocompatibility and/or provide specificity and function. The surface chemistry of nanoparticles can also be varied with the introduction of functional groups, thus providing specificity toward the immobilization of specific ligands. Silica (SiO2) is a promising material for the development of these platforms. Silica is generally considered biocompatible and nontoxic and has been applied in vivo.4,5 Silica nanoparticles (SiNPs) of size ranging from a few nanometers to hundreds of nanometers have been synthesized to make particles tailored to specific applications, including delivery into cells and tissues.6 The particle size is a © XXXX American Chemical Society

very important parameter in the design of such platforms because it defines the ability of a particle to permeate tissues and cells.7 In addition, particle topography at the nanometer scale also has an important effect on adsorbed protein structure and function.8,9 Traditional synthetic routes to spherical SiNPs of controlled size include the Stöber method10 or one of its variants11 and are based on the ammonia-catalyzed hydrolysis− condensation reactions of tetraethylorthosilicate (TEOS) with water, where the silica particle size in the range of 5−2000 nm is controlled by the initial concentration of ammonia. The presence of silanol groups makes a silica surface very versatile toward surface chemistry modifications and allows for a variety of surface functionalization and application.1,2,4,6 As an example, amino-modified SiNPs have been used as carriers for DNA and plasmids for potential use in diagnostic research and gene therapy.12 Modification of the silica surface with the Received: August 21, 2013 Revised: November 29, 2013

A

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Table 1. Physicochemical Properties of Silica with Different Particle Sizes and Surface Chemistries sample

size (nm) ± ± ± ±

SiO2-28 SiO2-82 SiO2-210 SiO2-500

28 82 210 470

CH3-SiO2-500 3-AP-SiO2-500

480 ± 3 380 ± 12

4 4 5 4

SSA (m2/g)a 243.2 65.0 20.8 8.5

± ± ± ±

1.6 0.3 0.4 0.5

pore volume (cm3/g)b Pristine Silica 0.278 0.334

pore size (nm)b

# OH/nm2c

ζ, pH 7.4 (mV)

12.2 32

4.3 4.5 6.5 6.3

−27 −38 −40 −43

4 5 3 5

3 2.4 2.3 2.2

NA NA

−30 ± 3 +60 ± 4

2.5 10

± ± ± ±

pzc

Functionalized 8.0 ± 0.4 12.1 ± 0.7

a

Specific surface area was determined using the Brunauer−Emmett−Teller (BET) method in the range of 0.05 < P/P0 < 0.35. bBJH desorption pore volume and size were calculated between 2 and 300 nm. cDetermined from TGA measurements (see Figure S5 of the Supporting Information). acid (TFA)/triisopropylsilane (TIS)/3,6-dioxa-1,8-octanedithiol (DODT)/water solution. Peptides were characterized by matrixassisted laser desorption ionization (MALDI) mass spectroscopy, and their purity (>85%) was assessed by high-performance liquid chromatography−mass spectroscopy (HPLC−MS) (see Figures S1− S3 of the Supporting Information). 2.2. Preparation and Characterization of Silica. All lyophilized silica particles were obtained by a modified Stöber method.10 Silica functionalization was obtained according to the method of Wu and colleagues.11 Unless otherwise specified, analysis and assays were carried out on lyophilized silica powders without further treatment. N2 adsorption−desorption isotherms were obtained on approximately 100 mg of powder using a Quantochrome Nova 3200e. Samples were degassed at 120 °C overnight prior to analysis. The surface area was determined from a five-point adsorption isotherm in the relative pressure range of 0.05−0.35 at 77.3 K, and porosity was evaluated from the desorption branch using the Barrett−Joyner−Halenda (BJH) method. Thermogravimetric analysis (TGA) was run on 3−7 mg of material using a TGA/SDTA 851 Mettler Toledo, heating from 30 to 800 °C at a 10 °C/min heating rate under a dry N2 flow. The OH density was calculated from the weight loss in the region of 150−800 °C using the following formula:19

introduction of additional/alternative surface functionality has been shown to affect the adsorption capacity of proteins, such as bovine serum albumin (BSA) and fibrinogen,13,14 as well as biomolecules of high molecular weight, such as γ globulins and other blood constituents.15 Binding interactions at the silica/aqueous interface are complex and depend upon several variables, including surface properties, the binding environment, and the nature of the binder. Small peptides (7−12-mer) identified through combinatorial approaches are often used as simple model biomolecules to study binding events at the molecular level.16−18 Important progress in the identification of key amino acids and motifs responsible for binding and insights about the binding mechanism and interactions during the binding event at the peptide/silica interface have recently been reported.16,18 Further, the synergistic application of experimental and computational approaches is providing an important contribution toward understanding and possibly prediction of interfacial phenomena.18 In particular, we have shown that, by controlling the binding environment (i.e., pH) or the surface hydrophilicity, the binding event at the silica/ peptide interface can be intentionally selected. These results show that, by knowing the physical and chemical properties of the interface components (i.e., the mineral surface and binder), insights into the mechanism of interaction can be achieved. Herein, to extend our understanding of the binding event at the silica/aqueous interface, we investigate the influence of (a) particle size and (b) surface functionalization, by systematically studying the binding behavior of a set of silica binders17 on a range of fully characterized Stöber materials of defined surface chemistry and particle size. The results will help us understand the effect that small changes in surface chemistry have on peptide affinity and will facilitate the design of interfaces where recognition−adsorption responses can be intentionally modulated, thus providing a valuable advancement toward the implementation of tunable interfaces in the biomedical and biotechnology fields.

# OH/nm 2 =

2[wt(Tf ) − wt(Ti)] NA MWH2O SBET

Methyl coverage was obtained using a similar approach. Attenuated total reflection Fourier transform infrared (FTIR) spectra in the midIR range (4000−800 cm−1) were obtained using a Perkin-Elmer Spectrum 100-FTIR spectrometer, and DRIFT FTIR spectra in the near IR (8000−4000 cm−1) were obtained using a Nicolet Magna IR 750 spectrometer, equipped with a DRIFT attachment. Spectra were recorded at a resolution of 4 cm−1 resolution, averaging 32 scans. ζ-Potential titrations were carried on 1 mg/mL silica suspensions in 10−3 M HCl and titrated with 10−3 M KOH. The ζ potentials were recorded on 1 mL of suspension in a capillary cell using a Malvern Zetasizer nano-S. Scanning electron microscopy (SEM) was run on freeze-dried samples attached to double-sided adhesive tape attached to SEM stubs and coated with carbon (Edwards, sputter coater S150B). The samples were then examined with a JEOL JSM-840A instrument operated at 20 kV. Transmission electron microscopy (TEM) was run on JEOL 2010 operating at 200 kV. Samples were prepared by evaporation of diluted suspensions onto carbon-coated copper grids. The particle size was assessed by averaging the size of a minimum of 50 particles from a SEM or TEM image. X-ray photoelectron spectroscopy (XPS) was performed on freezedried samples using a Surface Science M-probe XPS spectrometer with a Al Kα source (1486.6 eV) operated at a base pressure of 3 × 10−7 Pa using step sizes of 0. 065 eV for high-resolution XPS analysis and 1 eV for a general XPS survey. 2.3. Silica Binding Assay. Suspensions of silica (1 mg/mL) in phosphate-buffered saline (PBS; 100 mM phosphate and 150 mM NaCl) were sonicated for 1 h, and suitable amounts of peptide were added to achieve the desired initial peptide concentration and shaken

2. EXPERIMENTAL SECTION 2.1. Peptide Synthesis. The 7-mer peptides S1 (KLPGWSG), S2 (AFILPTG), and S3 (LDHSLHS) were identified by phage display.16 All peptide sequences were prepared in house by microwave-assisted solid-phase synthesis (Discover SPS microwave peptide synthesizer) as previously described.16 The general procedure employed Wang resins preloaded with the Fmoc-protected C-terminus residue. The peptide sequence was built by amino acid coupling using piperazine in N,Ndimethylformamide (DMF) as a deprotector and tetramethyluronium hexafluorophosphate (HBTU) and N,N-diisopropylethylamine (DIPEA) as an activator and an activator base, respectively. Cleavage of peptide from the resin was achieved by treatment in trifluoroacetic B

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Table 2. Properties of Isolated 7-mer Silica Binding Peptides16,18 number of side-chain functionalities S1 S2 S3

sequence

pI

net charge at pH 7

basic

acidic

amino

hydroxyl

average hydrophilicity

KLPGWSG AFILPTG LDHSLHS

10.1 6 6

1 0 −0.8

1 0 2

0 1 1

0 0 2

1 1 1

−0.3 −1 −0.1

vigorously. The suspension was left to equilibrate for 1 h at room temperature and then centrifuged (13 000 rpm for 5 min). The amount of peptide in solution was quantified by fluorimetric spectroscopy. In a typical assay, 20 μL of fluorescamine (5 mg/mL in acetone) was added to a 180 μL aliquot of the supernatant in a 96well plate and the fluorescence intensity was measured using a Tecan Spectrafluor XFLUOR4 plate reader equipped with a 360 nm excitation filter and a 465 nm absorption filter. The concentration of peptide was calculated using a peptide-specific calibration curve, and the amount of peptide adsorbed was calculated by difference with the initial peptide concentration. All assays were repeated 3 times to guarantee their reproducibility.

3. RESULTS AND DISCUSSION To guarantee consistency of surface properties and reproducibility of results throughout all stages of the present study, all SiNPs were sourced from a single batch and fully characterized. The physicochemical properties and surface chemistry of all nanoparticles synthesized were measured, and data were summarized in Table 1. The binding sequences S1 (KLPGWSG), S2 (AFILPTG), and S3 (LDHSLHS) were identified by biopanning as strong binders for 82 nm SiNPs.16 At binding assay conditions (ca. pH 7), the sequences differ in charge because of the different sidechain functionalities (Table 2). Differences in such properties define the binding mechanism to SiNPs and the interfacial interactions occurring at the silica−aqueous interface. For example, the cationic peptide S1 was shown to bind principally via electrostatic interactions; hydrophobic and hydrogenbonding interactions were responsible for peptide S2; and a combination of electrostatic and hydrophobic/hydrogenbonding interactions was observed for S3, depending upon binding conditions.16,18 3.1. Effect of the Silica Particle Size on Adsorption Properties. To study the effect of the particle size, a set of pristine silica spheres ranging between 28 and 500 nm in diameter (SiO2-28, SiO2-82, SiO2-210, and SiO2-500) were used (panels a−d of Figure 1). The BET surface area of the pristine silica samples ranged from 8 to 243 m2/g, showing the expected reciprocal relationship with the particle size. SiO2-210 and SiO2-500 are non-porous, while samples SiO2-28 and SiO282 have significant porosity (see Figure S4 of the Supporting Information) and are characterized by a well-developed mesoporous structure that results in high surface area and pore volumes. The shape of the isotherms (see Figure S4 of the Supporting Information) and the pore sizes (Table 1) indicate the interparticle origin of this porosity. The adsorption of each peptide as a function of the initial peptide concentration [pep]i at room temperature and pH 7.4 on silica spheres of size between 28 and 500 nm is reported in Figure 2. Because SiNPs of different size differ by surface area, the amount of adsorbed peptide is expressed per unit surface area. For SiNPs of any given size, peptide S1 showed a larger affinity, followed by S2 and S3, respectively, confirming the relative affinity of the peptides previously reported.16 Peptide adsorption on 82, 210, and 500 nm SiNPs increases with [pep]i

Figure 1. Electron microscopy of pristine and functionalized Stöber spheres: (a) SiO2-28, (b) SiO2-82, (c) SiO2-210, (d) SiO2-500, (e) CH3-SiO2-500, and (f) 3-AP-SiO2-500. Scale bars are 100 nm (a), 200 nm (b and c), and 500 nm (d−f).

up to ca. 1.5 mM, thereafter reaching saturation. The adsorption isotherms in [pep]i below the saturation point can be fitted using the Freundlich model (see Table S1 of the Supporting Information), indicating the formation of peptide multilayers on the silica surface16 up to a saturation level, with a total peptide uptake inversely proportional to the particle size. In the concentration range under study, no adsorption saturation is observed on the 28 nm particles, which still shows a Freundlich-type adsorption and has the lowest peptide uptake. For the three peptides at any [pep]i, the same general trend of increased adsorption capacity per unit surface area with increased silica particle size is observed. The adsorption trend observed is independent from peptide properties, such as charge or pI, in contrast to what was suggested by earlier qualitative studies,18 and is associated with an intrinsic property of the different sized SiNPs themselves. In other studies, an increase in adsorption capacity with particle size was observed for albumin and fibrinogen on a range of hydrophilic and hydrophobic SiNPs of diameter ranging between 15 and 165 nm.14 However, albumin and fibrinogen adsorption on silica was of Langmuir type, with the formation of a protein monolayer where the protein was oriented “end on” or “side on” depending upon particle curvature. The formation of lysozyme multilayers on 100 nm silica was also reported in another study,8 where a surface curvature effect was used to C

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 2. Adsorption isotherms of (a) S1 (KLPGWSG), (b) S2 (AFILPTG), and (c) S3 (LDHSLHS) on pristine SiNPs of different diameters. The adsorbed peptide was normalized to the BET surface area of the SiNPs.

Figure 3. ζ potential as function of pH for (a) pristine SiNPs and (b) functionalized SiNPs.

explain the results obtained. It should be noted that the spheres used in this current study cover a wider range of sizes than reported by others. Furthermore, the small peptide footprint (2 × 2 × 3.5 nm, with a surface footprint of ca. 7 nm2) is much smaller than that of, for example, lysozyme (4.5 × 3.5 × 3.5 nm, with a surface footprint of ca. 16 nm2), such that it is reasonable to assume that, at the peptide length scale, the curvature of the surfaces of all of the particle sizes can be considered negligible. Our detailed characterization of the surface chemistry of the SiNPs shows that the main difference between the particles is their surface charge density. Silica of different particle sizes, although having a similar point of zero charge (pzc), show differences in surface charge as a function of pH, as determined by ζ-potential titrations (Figure 3a). Variation of surface charge on silica in an aqueous environment is due to the protonation/ deprotonation equilibrium involving silanol groups (Figure 4a). Because protonation of silanol groups is considered negligible, SiNPs in water exhibit a negative electrical charge. The particle size is known to significantly affect the surface acidity of smallsized oxide nanoparticles in the quantum size range,21 by shifting the acid−base equilibrium to the right for small silica SiNPs (less than 10 nm).22 However, the difference in surface chemistry was found to be less significant for larger SiNPs (ca. 100 nm).24 The silica spheres used in this study are well outside the quantum size range and show an opposite trend of surface ionization.19,21 A general trend of a ζ potential increase with the particle size was observed: the ζ potential at pH 7.4 varied from −26 mV for the 28 nm sample to −56 mV for 500 nm silica, with a statistically significant difference between 28 nm particles and the other three samples. With an increasing particle size, a slight decrease of the pzc, from 3 to around 2.2, is observed. At

Figure 4. (a) Protonation/deprotonation equilibrium of silanol groups responsible for silica surface charge. In solution, any surface charge is balanced by the presence of counterions. (b) Surface chemistry of silica nanoparticles used in surface functionalization studies. (c) Silanol groups of different acidities present at the silica surface. The dotted line in c indicates hydrogen bonding.

pH 7.4, corresponding to the binding conditions, large particles have significantly higher ζ values. To understand the origin of the increased surface charge with the particle size, the properties of the silanol groups at the silica surface must be considered. At the silica surface exist silanol groups of different acidity:23 isolated (Q3), geminal (Q3), and vicinal (Q2) (Figure 4c). The relative distribution of Q3 and Q2 silanol groups depends upon the origin and pretreatment of the D

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 5. Effect of the surface chemistry on peptide uptake on silica: (a) S1 (KLPGWSG), (b) S2 (AFILPTG), and (c) S3 (LDHSLHS). All measurements were performed at pH 7.4.

sample24 and can affect the charge on silica surfaces having comparable total silanol density. We have estimated the OH density of our particles by TGA, considering the dehydroxylation weight loss between 200 and 800 °C.18 Thermograms are reported in Figure S5 of the Supporting Information. The silica powders contain between 4.3 and 6.5 OH/nm2, values in line with silanol densities obtained by TGA in previous studies25 and in good agreement with value of 4.6 OH/nm2 generally associated with the silanol density of fully hydroxylated silica.26 On the basis of this evidence, it is difficult to confidently explain the larger value obtained for the 210 and 500 nm diameter particles. This may be due to an effective higher silanol density of the large particles or can derive from the contribution of internal silanols to the weight loss in the temperature range of 200−800 °C. Internal silanols are not useful concerning surface properties and are expected to be more abundant in the larger particles than in the smaller particles.26 The observed increase in surface ionization with the particle size can arise from a higher population of isolated silanol groups compared to the smaller particles, where the more abundant population on the surface is represented by vicinal/ geminal groups.26 ζ-Potential measurements were correlated with higher mobility of counterions, confirming the stronger acidity of large particles. Improved adsorption for large particles can be attributed to their higher surface ionization.19 This is a straightforward explanation in the case of the cationic peptide S1 that binds predominantly by electrostatic interactions but not for peptides S2 and S3, for which a binding mechanism based on hydrophobic and hydrogen-bonding interactions, respectively, is anticipated. An explanation for this effect is a shift in the binding mechanism toward electrostatic interactions for adsorption of peptides on large silica spheres. Using a purely experimental approach to confirm this hypothesis would be a lengthy process. In addition, it is very difficult to obtain information at the molecular or atomic level with present experimental techniques. Accurate insight at the atomic scale could be obtained by computational studies, provided that a consistent force field is used to represent the interface. 3.2. Binding Experiments: Effect of the Surface Chemistry. Pristine silica particles of largest size (SiO2-500) were functionalized with methyl groups (CH3-SiO2-500) and 3aminopropyl groups (3-AP-SiO2-500) to obtain hydrophobic and positively charged surfaces, respectively. The aminopropylfunctionalized particles show a pzc of +10 because of protonation of the amino groups. The surface charge of methyl-modified materials, measured by ζ potential, does not show significant differences when compared to that of pristine spheres of similar particle size, showing a pzc of ca. 2.5 (Figure

3b). Porosity and surface area are similar to those of the pristine particle from which they derive. Amino-functionalized spheres are smaller than pristine silica because of partial dissolution of the spheres during the synthetic process (pH > 9). The binding of sequences S1 (KLPGWSG), S2 (AFILPTG), and S3 (LDHSLHS) on pristine and methyl- and 3-aminopropyl-functionalized silica spheres was studied as a function of the initial peptide concentration, [pep]i, at pH 7.4 (Figure 5). Silica samples of similar size (ca. 500 nm) were used to minimize the possible effect of the particle size itself. The adsorption isotherms are of Freundlich type, indicating the formation of peptide multilayers as previously reported, and do not reach saturation in the concentration range under study.16 The experimental data for the binding of both the cationic (S1) and anionic (S3) peptides fit with a binding model that involves mainly electrostatic interactions, where surface functionalization appears to disrupt the extent or efficiency of such interactions. The cationic peptide S1 shows higher affinity for the negatively charged pristine silica surface compared to the hydrophobic methyl-functionalized and the positively charged aminopropyl-functionalized surfaces. Conversely, a predictable increase in the adsorption of the negatively charged peptide S3 is observed on the amino-substituted surfaces, as compared to the silanol surface, while low affinity is shown for the methylated silica. It is perhaps surprising that the decrease in affinity for the modified surfaces is similar, despite the different nature of the surface functionalization. Peptide S1 (KLPGWSG) binds to unmodified silica preferentially through its positively charged amino terminus and lysine side chains.16,20 In the case of the aminopropylmodified silica, a charge effect can be considered responsible for the decrease in affinity. For the methyl-modified silica, although no significant difference in surface charge compared to the unmodified silica was observed (Figure 3b), the presence of methyl groups on the surface could affect the binding of the peptide by steric hindrance. Computational studies have shown that the anionic peptide S3 (LDHSLHS) binds pristine silica preferentially through hydrogen bonding involving serine and histidine residues with hydroxyl groups on the surface.20 Binding on methyl-modified silica is comparable to that of pristine silica up to [pep]i of 1 mM. A significant increase (>2fold) in S3 affinity for 3-aminopropyl-modified spheres compared to pristine particles is observed. This can clearly result from the contribution of electrostatic attraction to binding on the positively charged nanoparticles. In the case of the neutral peptide S2 (AFILPTG), the binding affinity decreases for the 3-aminopropyl-modified silica, while for the pristine and methyl-modified silicas, it is statistically equivalent at a high initial peptide concentration [pep]i. It should be noted E

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Table 3. Surface Composition (Atomic %) of Silica of Different Surface Chemistries, Confirming the Functionalization of Particle Surfaces C (atomic %) SiO2-500 CH3-SiO2-500 3-AP-SiO2-500

3.56 18.66 11.41

N (atomic %)

a

0.38

a

2.43

O (atomic %)

Si (atomic %)

66.51 54.55 63.32

29.56 25.95 22.82

a

The C amount is compatible with contamination from air adsorbents, and the N amount is compatible with residual traces of ammonia from the synthesis process.

surface binding response. Further studies are needed to understand if a higher surface coverage can be achieved with the method used herein or with other methods and what effect this would have on peptide adsorption. Overall, these results provide evidence of the impact that even very slight changes in the surface chemistry of an inorganic material can have on peptide affinity and binding. Such an influence is also likely to affect binder recognition and affinity during the biopanning process. In light of these results, it is clear that possible contamination of inorganic targets must be avoided or at least quantified and taken into account when analyzing the often varied pool of binders that result from biopanning experiments.

that, at a low initial peptide concentration [pep]i, the binder has a higher affinity for the hydrophobic methyl-modified surface compared to the more hydrophilic pristine surface. A large contribution to binding arises from hydrophobic interactions involving aliphatic and aromatic side chains present in the sequence that are enhanced by the presence of methyl groups on the silica surface, in a fashion similar to that shown for the binding of peptides to heat-treated silica.17 Surface functionalization is frequently used to introduce additional functionality or to allow for grafting of ligands. An accurate estimation of the topography of functionalization, in terms of surface coverage and distribution, is extremely challenging for nanoparticles.27 In previous studies,28,29 colorimetric, combustion elemental analysis, and TGA have been used to quantify surface functionalization; however, these methods are not specific for surface characterization. To the best of our knowledge, the effective surface functionalization achieved using the method herein reported has not been reported. Preliminary TGA of our particles strongly indicates functionalization well below that of a monolayer. The sample CH3-SiO2-500 shows a 2% weight loss at about 600 °C. Assuming that all weight loss derives from methyl groups at the surface, this weight loss corresponds to the presence of 0.3 CH3/nm2, corresponding to 14% of a monolayer. A weight loss resulting from the degradation of the aminopropyl functionalization was not detectable by TGA on sample 3-AP-SiO2-500 (see Figure S5 of the Supporting Information), indicating an even lower surface coverage. A degree of functionalization below that of a monolayer has been previously reported for mesoporous silica grafted in anhydrous conditions and was ascribed to the limited reactivity of surface silanols.29 XPS analysis was used to give a more precise estimation of the levels of C, N, O, and Si at the surfaces of the pristine and functionalized silica particles (see Figure S6 of the Supporting Information and Table 3). The content (atomic %) of C in CH3-SiO2-500 and N in 3-AP-SiO2-500 was used to estimate the degree of functionalization of the surface. The extent of functionalization was found to be less than 18.66 atomic % for methyl-modified particles and around 2.43 atomic % for the aminopropyl-modified particles. For peptide adsorption data collected for these particles, Figure 5 show that even the introduction of such a small amount of functionality on the surface of silica is sufficient to produce a significantly different response toward peptide uptake compared to that of pristine silica. This general trend is in agreement with changes observed for dye adsorption on modified silica nanospheres prepared by a similar method.11 The 2-fold affinity increase observed for S3 on the aminomodified NP is remarkable considering the low level of functionality present on the surface, which is evidently sufficient to shift the binding mechanism at the interface from largely that involving hydrogen-bonding interactions to a mode involving electrostatic interactions and generate a significant increase in

4. CONCLUSION In this study, we have used well-characterized batches of SiNPs to systematically investigate the effect of the particle size and surface functionalization on the adsorption affinity of a series of silica binders with different net charges at pH 7.4. We have estimated for the first time the surface coverage achieved for aminopropyl- and methyl-functionalized silica prepared according to Wu et al.11 For the binding studies, an effect of the silica particle size on peptide uptake per unit surface area was observed. It is proposed that a large particle size affects the absorption capacity as a result of increased surface ionization, causing a shift in the adsorption mechanism toward electrostatic interactions for the three peptides. A more detailed atomistic investigation of the binding event on such nanoparticles is needed. We show how a low degree of surface functionalization can be successfully used to modulate or select specific binding modes at the silica−peptide interface. A low surface coverage of hydrophobic functionalization (methyl-modified spheres) and polar functionalization (3-aminopropyl-modified spheres) was effective in disrupting electrostatic interactions at the silica− peptide interface. In addition, the 3-aminopropyl surface coverage, which was well below monolayer coverage, was sufficient to drastically increase the affinity for peptides capable of binding through polar interactions. Overall, we provide important insights into silica−biomolecule interfacial interactions and demonstrate how particle size and small changes in surface chemistry can define or modulate the affinity/binding properties at the (bio)molecule−silica interface.



ASSOCIATED CONTENT

S Supporting Information *

HPLC and MS traces for the three peptides used, adsorption isotherm fitting and parameters, N2 adsorption−desorption isotherms, thermograms, and correspondent XPS plots. This material is available free of charge via the Internet at http:// pubs.acs.org. F

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX

Langmuir



Article

(16) Puddu, V.; Perry, C. C. Peptide adsorption on SiNPs: Evidence of hydrophobic interactions. ACS Nano 2012, 6, 6356−6363. (17) Bachmann, M.; Goede, K.; Beck-Sickinger, A. G.; Grundmann, M.; Irback, A.; Janke, W. Microscopic mechanism of specific peptide adhesion to semiconductor substrates. Angew. Chem., Int. Ed. 2010, 49, 9530−9533. (18) Patwardhan, S. V.; Emami, F. S.; Berry, R. J.; Jones, S. E.; Naik, R. R; Deschaume, O.; Heinz, H.; Perry, C. C. Chemistry of aqueous silica nanoparticle surfaces and the mechanism of selective peptide adsorption. J. Am. Chem. Soc. 2012, 134, 6244−6256. (19) Kim, J. M.; Chang, S. M.; Kong, S. M.; Kim, K. S.; Kim, J.; Kim, W.S.. Control of hydroxyl group content in silica particle synthesized by the sol-precipitation process. Ceram. Int. 2009, 35, 1015−1019. (20) Emami, F. S.; Puddu, V.; Berry, R. J.; Naik, R. R.; Patwardhan, S. V.; Perry, C. C.; Heinz, H. Atomistic force field for silica: Influence of surface structure, particle size, and pH on interfacial chemistry and biomolecule adsorption. Manuscript in preparation. (21) Vayssieres, L. On the effect of nanoparticle size on water−oxide interfacial chemistry. J. Phys. Chem. C 2009, 113, 4733−4736. (22) Brown, M. A.l; Duyckaerts, N.; Beloqui Redondo, A.; Jordan, I.; Nolting, F.; Kleibert, A.; Ammann, M.; Wörner, A. J.; van Bokhoven, J. A.; Abbas, Z. Effect of surface charge density on the affinity of oxide nanoparticles for the vapor−water interface. Langmuir 2013, 29, 5023−5029. (23) Mendez, A.; Bosch, E.; Rose, M.; Neue, U. D. Comparison of the acidity of residual silanol groups in several liquid chromatography columns. J. Chromatogr., A 2003, 986, 33−44. (24) Zhuravlev, L. T. The surface chemistry of amorphous silica. Zhuravlev model. Colloids Surf., A 2000, 173, 1−38. (25) Ek, S.; Root, A.; Peussa, M.; Niinisto, L. Determination of the hydroxyl group content in silica by thermogravimetry and a comparison with 1H MAS NMR results. Thermochim. Acta 2001, 379, 201−212. (26) Ya Davydoav, V.; Kiselev, V.; Zhuravlew, L. T. Study of the surface and bulk hydroxyl groups of silica by infrared spectra and D2O exchange. Trans. Faraday Soc. 1964, 60, 2254−2264. (27) Van de Waterbeemd, M.; Sen, T.; Biagini, S.; Bruce, I. J. Surface functionalisation of magnetic nanoparticles: Quantification of surface to bulk amine density. Micro Nano Letters 2010, 5 (5), 282−285. (28) Moon, J. H.; Shin, J. W.; Kim, S. Y.; Park, J. W. Formation of uniform aminosilane layers: An imine formation to measure relative surface density of the amine group. Langmuir 1996, 12, 4621−4624. (29) Nieto, A.; Colilla, M.; Balas, F.; Vallet-Regı ́, M. Surface electrochemistry of mesoporous silicas as a key factor in the design of tailored delivery devices. Langmuir 2010, 26 (7), 5038−5049.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank the American Air Force Office of Scientific Research (AFOSR) (Grants FA9550-10-1-0024 and FA955013-1-0040) and the Engineering and Physical Sciences Research Council (EPSRC) (EP/E048439/1) for funding this work. The authors are grateful to D. J. Belton and G. J. Hickman for kindly providing silica samples. J. M. Slocik (Air Force Research Laboratory, RXBN) is acknowledged for performing XPS analyses.



REFERENCES

(1) Li, Z.; Barnes, J. C.; Bosoy, A.; Stoddart, J. F.; Zink, J. I. Mesoporous silica nanoparticles in biomedical applications. Chem. Soc. Rev. 2012, 41, 2590−2605. (2) Schulz, A.; McDonagh, C.. Intracellular sensing and cell diagnostics using fluorescent silica nanoparticles. Soft Matter 2012, 8, 2579−2583. (3) Ma, A. K.; Zhang, H. X.; Liu, K. H.; Wei, T. J.; Zou, W. C.; Liang, X. G. Gold nanoparticles functionalized with therapeutic and targeted peptides for cancer treatment. Biomaterials 2012, 33, 1180−1189. (4) Wu, S. H.; Lin, Y. S.; Hung, Y.; Chou, Y. H.; Hsu, Y. H.; Chang, C.; Mou, C. Y. Multifunctional mesoporous SiNPs for intracellular labeling and animal magnetic resonance imaging studies. Chem. Biol. Chem. 2008, 9, 53−57. (5) Park, J. H.; Gu, L.; von Maltzahn, G.; Ruoslahti, E.; Bhatia, S. N; Sailor, M. J.. Biodegradable luminescent porous silicon nanoparticles for in vivo applications. Nat. Mater. 2009, 8, 331−336. (6) Qianjun, H.; Jianlin, S. Mesoporous silica nanoparticle based nano drug delivery systems: Synthesis, controlled drug release and delivery, pharmacokinetics and biocompatibility. J. Mater. Chem. 2011, 21, 5845−5855. (7) Fang, L.; Si-Han, W.; Yann, H.; Chung-Yuan, M. Size effect on cell uptake in well-suspended, uniform mesoporous silica nanoparticles. Small 2009, 5, 1408−1413. (8) Vertegel, A. A.; Siegel, R. W.; Dordick, J. S. Silica nanoparticle size influences the structure and enzymatic activity of adsorbed lysozyme. Langmuir 2004, 20, 6800−6807. (9) Roach, P.; Farrar, D.; Perry, C. C. Interpretation of protein adsorption: Surface-induced conformational changes. J. Am. Chem. Soc. 2005, 127, 8168−8173. (10) Stöber, W.; Fink, A.. Controlled growth of monodisperse silica spheres in the micron size range. J. Colloid Interface Sci. 1968, 26, 62− 69. (11) Wu, Z.; Xiang, H.; Taehoon, K.; Chun, M. S.; Lee, K. Surface properties of submicrometer silica spheres modified with aminopropyltriethoxysilane and phenyltriethoxysilane. J. Colloid Interface Sci. 2006, 304, 119−124. (12) He, X.; Wang, K.; Tan, W.; Liu, B.; Lin, X.; He, C.; Li, D.; Huang, S.; Li, J.. Bioconjugated nanoparticles for DNA protection from cleavage. J. Am. Chem. Soc. 2003, 125, 7168−7169. (13) Roach, P.; Farrar, D.; Perry, C. C. Surface tailoring for controlled protein adsorption: Effect of topography at the nanometer scale and chemistry. J. Am. Chem. Soc. 2006, 128, 3939−3945. (14) Roach, P.; Shirtcliffe, N. J.; Farrar, D.; Perry, C. C. Quantification of surface-bound proteins by fluorometric assay: Comparison with quartz crystal microbalance and amido black assay. J. Phys. Chem. B 2006, 110, 20572−20579. (15) Yildirim, A.; Ozgur, E.; Bayindir, M. Impact of mesoporous silica nanoparticle surface functionality on hemolytic activity, thrombogenicity and non-specific protein adsorption. J. Mater. Chem. B 2013, 1, 1909−1920. G

dx.doi.org/10.1021/la403242f | Langmuir XXXX, XXX, XXX−XXX