SANS and UV–vis Spectroscopy Studies of Resultant Structure from

Jun 27, 2011 - The measurements were performed on fixed concentration (1 wt %) of nanoparticles and varying concentration of protein in the range 0 to...
0 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/Langmuir

SANS and UVvis Spectroscopy Studies of Resultant Structure from Lysozyme Adsorption on Silica Nanoparticles Sugam Kumar,† Vinod K. Aswal,†,* and Joachim Kohlbrecher‡ † ‡

Solid State Physics Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India Laboratory for Neutron Scattering, Paul Scherrer Institut, CH-5232 PSI Villigen, Switzerland ABSTRACT: The interaction of lysozyme protein (M.W. 14.7 kD) with two sizes of silica nanoparticles (16 and 25 nm) has been examined in aqueous solution using UVvis spectroscopy and small-angle neutron scattering (SANS). The measurements were performed on fixed concentration (1 wt %) of nanoparticles and varying concentration of protein in the range 0 to 2 wt %. The adsorption isotherm as obtained using UVvis spectroscopy suggests strong interaction of the two components and shows an exponential behavior. The saturation values of adsorption are found to be around 90 and 270 protein molecules per particle for 16 and 25 nm sized nanoparticles, respectively. The adsorption of protein on nanoparticles leads to the aggregation of particles and these structures have been studied by SANS. The aggregates are characterized by fractal structure coexisting with unaggregated particles at low protein concentrations and free proteins at higher protein concentrations. Further, contrast variation SANS measurements have been carried out to differentiate the adsorbed and free protein in these systems.

’ INTRODUCTION Small size but large surface-to-volume ratio of nanoparticles and their surface-dependent properties make them unique and distinct compare to the bulk of that material, which leads to a wide range of potential applications of these systems in the area of nanobiotechnology, optics, and electronics. Many of these applications depend on interfacing the nanoparticles with macromoleclues such as surfactant, polymer and protein.1,2 For example, nanoparticle-surfactant interactions are important in many industrial and technical applications related with enhanced dispersion stability, chemical mechanical polishing, and design of functional nanomaterials.35 The nanoparticles functionalized with polymers have been utilized in developing protein sensor arrays.6 Also nanoparticle interaction with protein enables them to be used in the emerging field of nanobiotechnology (nanomedicine, drug delivery, and biosensors) as the nanoparticles having sizes comparable to that of living cells and can access to and operate within the cell.79 The nanoparticle and macromolecule can interact through various interactions such as covalent bonding, electrostatic forces, hydrogen bonding etc. depending on the system of interest.5,1012 These interactions are used to tune the system properties as required for different applications. In particular, when the two components (nanoparticles and macromolecules) are oppositely charged, the electrostatic interaction become specifically important for such tuning.1315 Also the electrostatic forces lead to the many nonspecific associations of macro molecules especially relevant in biological systems (e.g., chromatin and protein complexes) together with the other prominent processes such as formation of r 2011 American Chemical Society

polyelectrolyte multilayer and coassembly of double hydrophilic block copolymers in the presence of oppositely charged species.12,1618 It has been recognized that the attractive interaction between macromolecules and nanoparticles give rise to new types of versatile hybrid structures combining the special properties of both the organic and inorganic worlds that have numerous advantages such as filtration, care products, bioelectronics, nanomedicine, and nanocomposite technology.12,18,19 While there are number of structural studies available for the nanoparticle and polymer12,19,20 or surfactant5,21,22 composites, no such study has been carried on the nanoparticleprotein complexes. Proteins are charged macromolecules and their function depends on the native folded structure. Therefore, it is possible during their interaction with nanoparticles that proteins are deformed and can have the conformations which are very different than those from the native structure. In this regard, various techniques have been used to examine amount of adsorption and the functionality of adsorbed protein on nanoparticle.23 In particular, UVvis spectroscopy has been used to obtain the adsorption isotherms for the protein adsorption on nanoparticle surface.24,25 However, techniques like circular dichroism (CD), nuclear magnetic resonance (NMR), and Raman spectroscopy are being utilized to observe the conformational changes occurred in the native protein structure as a result of their Received: April 8, 2011 Revised: June 27, 2011 Published: June 27, 2011 10167

dx.doi.org/10.1021/la201291k | Langmuir 2011, 27, 10167–10173

Langmuir interaction with nanoparticle.23,26,27 These studies clearly show that the amount of adsorption and structure perturbation strongly depend on the properties of both nanoparticle2831 (e.g., size, shape, surface roughness, and charge density) and solution behavior of protein which can be varied by the concentration, temperature, pH, and on addition of salt.11,24,32 Small-angle neutron scattering (SANS) is a useful technique to study nanostructured materials and in particular has unique advantages to characterize multicomponent systems such as nanoparticleprotein complexes.33,34 This arises because of the easy possibility to vary the contrast by either deuterating the solute or solvent in order to visualize the role of each component in the complex multicomponent system.35,36 In the present work, emphasis is on characterizing the structures formed between the silica nanoparticles and lysozyme protein through electrostatic complexion. The measurements are carried out at pH 7 at which the two components are oppositely charged. Lysozyme is a small globular protein (M.W. 14.7 kD) and one of the most commonly studied in the literature.37 SANS provides the information on the amount of adsorption and the resultant structures formed in these system. UVvis spectroscopy has also been carried out to measure the adsorption isotherms prior to SANS studies.

’ EXPERIMENTAL SECTION Electrostatically stabilized colloidal suspension of silica nanoparticles (Ludox LS30 and TM50) and hen egg protein lysozyme (catalogue no. 62970) were purchased from Sigma-Aldrich and Fluka, respectively. The adsorption isotherms of protein interaction with silica nanoparticles in aqueous solution were studied using a nanodrop spectrophotometer ND 1000. The instrument is based on surface retention technology utilizing the surface tension to hold the sample. A pulsed Xenon flash lamp is used as a source to cover the spectrum range from 220 to 750 nm and the light coming through the sample is analyzed by CCD arrays. Samples for SANS experiments were prepared by dissolving weighted amount of nanoparticle and protein in 20 mM phosphate buffer at pH 7 prepared in D2O. The measured zeta potential values of LS30 and TM50 silica nanoparticles are 38 mV and 30 mV, respectively. However, the reported value of the zeta potential for lysozyme protein is +4 mV and is known to have a net positive charge of 8 e at pH 7.38 In neutron scattering experiments, use of D2O over H2O is preferred because of high contrast for hydrogenous samples in D2O. Some samples were also prepared in D2O/ H2O mixed solvent to vary the contrast for silica particles as well as proteins. Small-angle neutron scattering experiments were performed at SANS-I facility, Swiss Spallation Neutron Source SINQ, Paul Scherrer Institut, Switzerland.39 The wavelength (λ) of neutron beam used was 6 Å and the scattered neutrons from samples were detected using two-dimensional 96 cm96 cm detector. Data were collected at two sample-to-detector distances 2 and 8 m to cover a wave vector transfer (Q = 4πsin(θ/2)/λ, where θ is scattering angle) range of 0.008 to 0.3 Å1. All of the measurements were carried out for fixed concentration (1 wt %) of silica nanoparticles and varying the concentration of protein in the range 0 to 2 wt %. The freshly prepared samples were held in HELLMA quartz cells having thickness 2 mm and temperature kept constant at 30 C during the measurements. There was no precipitation observed during the measurements. The data were corrected and normalized to absolute scale using standard procedure.

’ SANS ANALYSIS In SANS experiments, one measures the coherent differential scattering cross-section per unit volume (dΣ/dΩ) as a function

ARTICLE

of Q. In the case of monodispersed particles dispersed in a medium, it can be written as follows:40,41 dΣ ðQ Þ ¼ nV 2 ðFp  Fs Þ2 PðQ ÞSðQ Þ þ B dΩ

ð1Þ

where n is the number density and V is particle volume. Fp and Fs are scattering length densities of particles and solvent respectively. P(Q) is intraparticle structure factor and S(Q) is interparticle structure factor. B is a constant term representing incoherent background. P(Q) depends on shape and size of the particle and is the square of single particle form factor F(Q) as given by the following: PðQ Þ ¼ ÆjFðQ Þj2 æ

ð2Þ

For a spherical particle of radius R, F(Q) is given by the following: FðQ Þ ¼

3fsinðQRÞ  ðQRÞcosðQRÞg ðQRÞ3

ð3Þ

S(Q) correlates particles present in the system and it is the Fourier transform of the radial distribution function g(r) for the mass centers of the particle. For diluted samples, S(Q) ≈ 1. For polydispersed systems, dΣ/dΩ in eq 1 is modified as follows:42 Z dΣ dΣ ðQ Þ ¼ ðQ , RÞ f ðRÞdR þ B ð4Þ dΩ dΩ where f(R) is the size distribution and usually accounted by Schultz distribution as given by the following:       Z þ 1 Zþ1 Z Z þ 1 1 f ðRÞ ¼ R exp  R Rm Rm ΓðZ þ 1Þ ð5Þ where Rm and Z are the mean value and width of distribution, respectively and Γrepresents to the Gamma function. The polydispersity of this distribution is given by σ = 1/(Z+1)1/2. In the case of particle aggregation as characterized by fractal structure, the scattering cross-section can be expressed as follows:43 dΣ ðQ Þ ¼ na Vp2 ðFp  Fs Þ2 PðQ ÞSf ðQ Þ þ B dΩ

ð6Þ

where na and Vp are the number density and volume of individual scatterer in the aggregates. P(Q) is the intraparticle structure factor of the building block in aggregated structure as given by eq 2 and Sf (Q) is the structure factor. Sf (Q) for mass fractal is given by the following:43 Sf ðQ Þ ¼ 1 þ

1 " ðQRÞD

tan1 ðQ ξÞg

DΓðD  1Þ #ðD  1Þ=2 sinfðD  1Þ 1 1 þ ðQ ξÞ2 ð7Þ

where ξ signifies the maximum length up to which fractal microstructure exists, R is the size of building block and D is the fractal dimension. The scattering length densities of silica nanoparticles, lysozyme protein, D2O and H2O used are 3.8  1010, 0.39  1010, 6.4  1010and 0.56  1010/cm2, respectively. Throughout the 10168

dx.doi.org/10.1021/la201291k |Langmuir 2011, 27, 10167–10173

Langmuir

ARTICLE

Figure 2. Lysozyme protein adsorption isotherm in 1 wt % LS30 silica nanoparticle solution. The variations of adsorbed protein concentration as well as percent adsorbed protein are plotted as a function of protein concentration. Inset shows the protein concentration dependent adsorption of protein molecules per nanoparticle.

Figure 1. SANS data of 1 wt % silica nanoparticles (LS30 and TM50) and lysozyme protein solutions.

Table 1. Fitted Parameters of 1 wt % Silica Nanoparticles (LS30 and TM50) and 1 wt % Lysozyme Protein Solutions (a) silica nanoparticles sample

mean radius Rm (Å)

polydispersity σ

LS30

80.0 ( 1.0

0.15 ( 0.02

TM50

125.5 ( 2.0

0.15 ( 0.02

(b) lysozyme protein semimajor

semiminor axis

equivalent radius

sample

axis a (Å)

(b = c) (Å)

Re (Å)

lysozyme

24.0 ( 0.5

13.5 ( 0.1

16.3 ( 0.2

data analysis corrections were made for instrumental smearing where calculated scattering profiles are smeared by the appropriate resolution function to compare with the measured data. The parameters in the analysis were optimized by means of nonlinear least-squares fitting program.44

’ RESULTS AND DISCUSSION SANS data of 1 wt % pure silica nanoparticles (LS30 and TM50) and lysozyme protein in aqueous solution are shown in Figure 1. Both the data of nanoparticles and lysozyme show a monotonically decreasing profile as a function of Q. At 1 wt % (0.4 vol %) concentration, the interparticle structure factor S(Q) contribution can be neglected because interparticle distance among the particles at this concentration is much larger compare to their sizes where the interference among the particles are significantly small. This was further confirmed by the scaling of the SANS data measured at lower concentrations to 1 wt % data. Therefore the scattering is mostly governed by intraparticle structure factor P(Q).45 The hump in the scattering profiles at Q values about 0.07 and 0.04 Å1 for LS30 and TM50 nanoparticles, respectively are as a result of P(Q) oscillations of polydispersed

systems. Silica nanoparticles are modeled by the polydispersed spheres using eq 4 and the fitted parameters of the analysis are given in table 1. The lysozyme is a small globular protein and is fitted with prolate ellipsoidal shape.46 The scattering buildup in lysozyme system at low Q values is attributed to the existence of some permanent aggregates.47 The calculated values of semimajor and semiminor axes are found to be 24.0 ( 0.5 and 13.5 ( 0.1 Å, respectively. The reason for choosing LS30 and TM50 systems had been that they consist of nanoparticles having significantly different sizes as determined in table 1. Also the sizes of nanoparticles are significantly larger than that of the proteins to provide enough particle surface area for the interaction of protein. The adsorption isotherm of lysozyme protein on LS30 silica nanoparticles (2Rm = 16 ( 0.2 nm) as obtained by UVvis spectroscopy is depicted in Figure 2. The samples for these experiments were prepared by mixing the fixed concentration of 1 wt % silica nanoparticles and varying concentrations of lysozyme protein in aqueous solution. These samples are filled in cuvettes and centrifuged to separate free protein if any from that of the protein adsorbed on the nanoparticles. After the first run, the supernatants were carefully removed and kept in fresh cuvettes. The same procedure is repeated couple of times to improve the separation process. The UVvis spectrum of the lysozyme shows a peak at about 280 nm due to the absorption of the incident light by protein.48 The concentration of the free protein in the sample is calculated by measuring the ratio of this absorbance of the supernatants to the corresponding pure protein solution.24 It is seen in Figure 2 that at low protein concentrations the amount of the adsorbed protein increases with the increase in concentration and saturates at high protein concentrations. These data are fitted with an exponential rise of the adsorbed protein (A) as a function of protein concentration (C) as given by A = A0[1  expkC], where A0 is the saturation value and k is the adsorption coefficient. The values of A0 and k are found to be 0.50 ( 0.02 wt % protein and 2.0 ( 0.3 per wt% protein, respectively. Based on these data, we have also calculated the concentration dependent propensity (A/C) of protein adsorption on nanoparticles (Figure 2). The results show while the adsorption of protein is very high at low protein concentrations (e.g., 95% at 0.05 wt %) it significantly decreases at higher protein concentrations (e.g., 20% at 2 wt %).26 The inset of Figure 2 shows the variation of calculated number of adsorbed protein molecules (proportional to the adsorbed protein concentration) on individual nanoparticle with protein concentration. 10169

dx.doi.org/10.1021/la201291k |Langmuir 2011, 27, 10167–10173

Langmuir

Figure 3. SANS data of 1 wt % LS30 silica nanoparticle system on addition with (a) lower protein concentration (0.0 to 0.1 wt %) and (b) higher protein concentration (0.2 to 2 wt %).

Figure 3 shows the SANS data of 1 wt % silica nanoparticles with varying concentration of lysozyme. The data are divided into two sets: (i) low protein concentration in Figure 3(a) and (ii) high protein concentration in Figure 3(b) regimes. The first data set is considered in the protein concentration range from 0 to 0.1 wt % which shows the rise in scattering toward linearity on loglog scale in the low Q region and no significant change at high Q region on addition of protein. However, all of the data in the second set (protein concentration > 0.2 wt %) show the linearity on loglog scale with no significant change in the low Q region on further addition of protein. Moreover, in data set II there is distinct buildup in scattering profile at higher Q region with increasing protein concentration [Figure 3(b)]. The large rise in the scattering intensity in the low Q region in Figure 3 (a) cannot be explained based solely on the adsorption of protein on individual nanoparticles (coreshell structure) because of the low scattering from proteins (Figure 1). However, these data can be explained by the aggregation of the particles arising as a result of neutralization of charge on the nanoparticle by the adsorption of protein.23 The aggregates are fitted with the fractal structure as indicated by linearity in the low Q region at loglog scale.49 The analysis shows that in the first data set protein concentration is not enough to aggregate all the particles and therefore particle aggregates coexist with the free particles. The fitting involves the resultant scattering cross-section governed by the sum of two contributions from fractal aggregates [eq 6] and nonaggregated nanoparticles [eq 1]. The aggregates are found to be fractal structures having fractal dimensions about 2.4 ( 0.2 and the number fraction of aggregated particles increases with increasing protein concentration. The fitted parameters are given in Table 2. It is also found that all of the nanoparticles get aggregated for protein concentrations at 0.1 wt % and beyond this concentration. It is believed that the charge neutralization by protein

ARTICLE

adsorption on the silica nanoparticles approaches the value that all particles can aggregate for these protein concentrations. The isoelectric point of the mixed 1 wt % silica with varying lysozyme is obtained at its 0.25 wt % concentration. This is typically the value before which all of the nanoparticles in SANS have been found to be aggregated. Moreover, protein molecules can still be adsorbed on the nanoparticles beyond this concentration (0.25 wt %) up to the value of their saturation at around 0.5 wt %. The data of second set consist of all particles aggregates as there is no significant change in the data in the low Q region. The buildup observed at high Q region is because of free proteins which are not adsorbed on the nanoparticles. In this case, the data are analyzed by combing eqs 1 and 6 for free proteins and aggregates, respectively. Table 2 gives the free protein concentration in these systems. The nanoparticleprotein interaction in the aggregated structure is characterized by two sizes, namely the building block size and the effective size. These two sizes are differentiated in fitting [eq 6] for P(Q) to calculate effective particle size [eq 2] and S(Q) to calculate building block size [eq 7]. The building block radius and effective particle radius are found to be same (87 ( 1.7 Å), whereas their values are significantly larger than the radius of the individual silica nanoparticle (80 ( 1.0 Å). This increase is expected as the attraction of nanoparticles is mediated by the presence of oppositely charged protein molecules between them. The difference in the size of the building block and the nanoparticle (7 Å) is significantly less than that of the folded size of the protein molecule (Table 1), which could be as a result of disruption in the folded structure of protein on interaction with silica nanoparticle.23,50 Further any features corresponding to the overall size of the aggregates (low Q cutoff) are not observed within Q range of our measurements. However, it is clear from eq 6 that the scattering beyond low cutoff will be independent of the aggregate size. A fractal dimension 2.4 ( 0.2 in three dimensional Euclidean space indicates a diffusion limited aggregate (DLA) type of fractal morphology of the aggregates. 51,52 These fractal structures are highly branched and usually formed when the density of particles is quite low and the repulsive forces are relatively weak, as is the case in the present system. It may be mentioned that the results of SANS are consistent with that of UVvis spectroscopy where first and second SANS data sets correspond to the region I and II in UVvis spectroscopy, respectively. The first SANS data set and region I in UVvis spectroscopy corroborate to the high protein adsorption, whereas a second SANS data set and region II in UVvis spectroscopy confirms the existence of free protein in the system. The coexistence of fractal structure and the free protein at higher protein concentrations (data set 2 in Figure 3) is further confirmed by the contrast matching SANS data of the free protein to the solvent (Figure 4) by making the sample in the mixed (H2O/D2O) solvent containing 86 vol % of H2O. It is interesting to note that all of the data in Figure 4 are similar irrespective of the protein concentration in these systems. Since protein is contrast-matched, the scattering in Figure 4 is expected to be only from the fractal structure of the particle aggregates. The fact that the identical data are obtained in these systems with varying protein concentration confirms that all of the systems have the same fractal structure and the addition of protein leads to the increase in the number of free proteins consistent with that of increase in the scattering intensity at higher Q values in Figure 3(b). The calculated fractal dimension and building block radius (Table 3) have the same values as obtained without 10170

dx.doi.org/10.1021/la201291k |Langmuir 2011, 27, 10167–10173

Langmuir

ARTICLE

Table 2. Fitted Parameters of Interaction of 1 wt % LS30 Silica Nanoparticle System with Varying Lysozyme Protein Concentration (a) low protein concentration regime where the nanoparticle aggregates coexist with unaggregated nanoparticles concentration C (wt%)

building block radius Rb (Å) 80.0 ( 1.0

80.0 ( 1.0

0

2.4 ( 0.2

87.0 ( 1.7

87.0 ( 1.6

23

0.0 0.01

fraction of aggregated nanoparticles ϕanp (%)

fractal dimension D

effective particle radius Re (Å)

0.02

2.4 ( 0.2

86.0 ( 1.7

86.0 ( 1.7

43

0.05 0.1

2.4 ( 0.2 2.4 ( 0.2

86.0 ( 1.6 87.0 ( 1.7

86.0 ( 1.6 87.0 ( 1.7

83 100

(b) high protein concentration regime where nanoparticle aggregates coexist with excess free proteins. concentration C (wt%)

fractal dimension D

building block radius Rb (Å)

effective particle radius Re (Å)

fraction of free protein ϕfp (%)

0.2

2.4 ( 0.2

86.0 ( 1.9

86.0 ( 1.7

42

0.4

2.4 ( 0.2

87.0 ( 1.7

87.0 ( 1.6

48

1.0

2.4 ( 0.2

87.0 ( 1.7

87.0 ( 1.7

60

1.2

2.4 ( 0.2

87.0 ( 1.9

87.0 ( 1.9

69

1.6

2.4 ( 0.2

86.0 ( 1.7

86.0 ( 1.7

75

2.0

2.4 ( 0.2

87.0 ( 1.8

87.0 ( 1.8

82

Table 3. Fitted Parameters of Interaction of 1 wt % LS30 Silica Nanoparticle System with Lysozyme Protein in Which the Protein is Contrast-Matched to the Solvent concentration C (wt%)

Figure 4. SANS data of 1 wt % LS30 silica nanoparticle system (LS30) with varying concentration of lysozyme in which protein is contrastmatched to the solvent.

contrast matching protein system (Table 2). However, the effective particle radius is reduced from 87 ( 1.7 Å to the radius of the silica nanoparticle (80 ( 1.0 Å) since contrast-matched adsorbed proteins are not seen. To examine the effect of particle size on protein adsorption, we compare the adsorption isotherms of LS30 (16 nm) with that of larger particle TM50 (25 nm). The data are compared for 1 wt % nanoparticles concentration and are shown in Figure 5. These data show the similar trends for both the sizes of the particles with differences in their adsorption coefficient and saturation values. The adsorption coefficient and saturation values are increased from 2.0 ( 0.3 to 2.2 ( 0.3 per protein wt % and 90 ( 4 to 270 ( 10 protein molecules per particle, respectively by increasing the size of the particles from 16 to 25 nm. The adsorption coefficient and saturation values for given types of particle and protein is believed to depend on the parameters such as curvature and surface area of the particle.24,27 The combined effect of these parameters shows a small increase in the value of adsorption coefficient whereas the saturation value (number of protein molecules adsorbed per particle) increases significantly mostly because of an increase in the surface area of the particle. Inset of Figure 5 shows the variation of surface number density for

fractal dimension D

building block radius Rb (Å)

effective particle radius Re (Å)

0.2

2.4 ( 0.2

86.0 ( 1.6

80.0 ( 1.0

0.4

2.4 ( 0.2

87.0 ( 1.7

80.0 ( 0.8

1.0

2.4 ( 0.2

86.0 ( 1.6

80.0 ( 1.0

1.6

2.4 ( 0.2

87.0 ( 1.7

80.0 ( 0.9

2.0

2.4 ( 0.2

87.0 ( 1.7

80.0 ( 1.0

two sizes of silica nanoparticles. It is clear that the surface number density is higher for the larger particles having smaller curvature. Figure 6 shows the SANS data of the 1 wt % TM50 silica nanoparticle system on addition of the varying concentration of lysozyme. These data also show two important contributions similar to that discussed in Figure 3, first one in the low Q region coming from the fractal aggregates of the particles and second one that from free proteins in the high Q region. The scattering in the low Q region is dominated by the silica particles and the fact that most of these particles get saturated by the protein adsorption around 0.4 wt %, all of the SANS data measured with protein concentration beyond this value show the almost same structure of aggregated particles. The scattering from free proteins in the high Q region as expected increases with the protein concentration. The fitted parameters from these systems considering the coexistence of fractal particle aggregates with free proteins are given in Table 4. The particle aggregates have the fractal dimension of 2.4 ( 0.2 consisting of building block radius and effective particle radius (particle + adsorbed protein) 133 ( 4.0 Å. The value of fractal dimension is found to be similar for two sizes of particles as a result of non site-specific aggregation of these systems. However, the increase in the size of building block is consistent with the increase in the size of the particle. We have seen that the percentage of adsorbed protein decreases with an increase in the protein concentration. 10171

dx.doi.org/10.1021/la201291k |Langmuir 2011, 27, 10167–10173

Langmuir

ARTICLE

Table 4. Fitted Parameters of Interaction of 1 wt % TM50 with Varying Lysozyme Protein Concentration building

effective

fraction of

concentration

fractal

block radius

particle radius

free protein

C (wt%)

dimension D

Rb (Å)

Re (Å)

ϕfp (%)

0.5

2.4 ( 0.2

133 ( 4.0

132 ( 4.0

45

1.0

2.4 ( 0.2

132 ( 3.8

133 ( 4.0

59

2.0

2.4 ( 0.2

133 ( 3.6

133 ( 3.7

85

Figure 5. Comparison of lysozyme protein adsorption isotherms for the two sizes of silica nanoparticles. Inset shows the variation of surface number density of the adsorbed protein in these systems. Figure 7. Square root of the scattering intensity after correcting for incoherent background for 1 wt % TM50 silica nanoparticle system as a function of % H2O in the mixed (H2O/D2O) solvent.

Figure 6. SANS data of 1 wt % TM50 silica nanoparticle system with varying concentration of lysozyme protein.

For differentiating the adsorbed protein and the free protein, the scattering data were collected from the nanoparticleprotein systems where silica nanoparticles are contrast-matched. It was done by preparing the sample in a mixed solvent of H2O and D2O. To know first the exact composition of mixed solvent that is contrast-matched to the silica nanoparticles the SANS data were taken from pure nanoparticle system with varying the solvent composition. The linear plot of the square root of scattered intensity with solvent composition gives the contrast match point corresponding to zero scattering intensity as shown Figure 7.53 It is found that silica nanoparticles are contrast-matched for 37 vol % of H2O in the H2O/D2O mixed solvent.36,54 SANS data of 1 wt % TM50 silica nanoparticle with varying lysozyme protein in contrast-matched solvent are shown in Figure 7. All of the scattering data show a buildup in the low Q region as a result of scattering from the protein adsorbed on the nanoparticles. Since the adsorbed protein layer is formed at a much larger length scale (size of nanoparticle) than the size of individual protein, the scattering from such shell structure (core is contrast-matched) is expected to be in the low Q region. Unfortunately, these data are superimposed on high incoherent background of the mixed solvent (37 vol % H2O) and also the contrast for the protein becomes weak. The data are therefore flat at high Q region up to the protein concentration of 2 wt %. This limits us to perform the quantitative analysis of all of these data. However, there are clearly two buildups for the data of high protein concentrations,

Figure 8. SANS data of 1 wt % TM50 silica nanoparticle system with varying lysozyme protein concentration in which the nanoparticles are contrast-matched to the solvent. Inset shows the fitted data for one of the systems (1 wt % silica nanoparticles with 10 wt % lysozyme).

first one rise of scattering in the low Q region corresponding to the adsorbed protein and second one emergence of correlation peak arising from the electrostatic repulsion between molecules at Q ≈ 0.1 Å1 relating to the free protein. This model is further verified by fitting the data of 1 wt % contrast-matched silica with 10 wt % lysozyme as shown in the inset of Figure 8. Contrast variation SANS measurements thus confirm to the coexistence of two populations (adsorbed and free proteins) on interaction of proteins with nanoparticles.

’ CONCLUSIONS Small-angle neutron scattering along with UVvis spectroscopy has been used to study resultant structures formed from lysozyme protein adsorption on two different sized silica nanoparticles. The studies were carried out on the nanoparticle systems having sizes much larger than the protein size. The protein adsorption on nanoparticle shows an exponential behavior with adsorption coefficient and saturation values increasing with the size of the nanoparticle. The adsorption of protein leads to the aggregation of nanoparticles which are characterized by the 10172

dx.doi.org/10.1021/la201291k |Langmuir 2011, 27, 10167–10173

Langmuir fractal structure. The nanoparticle aggregates coexist with unaggregated particles at low protein concentrations whereas at high protein concentrations the excess free proteins are found with the nanoparticle aggregates. It is also found that the native protein structure is disturbed on their adsorption on nanoparticles.

’ AUTHOR INFORMATION Corresponding Author

*Phone: +91 22 25594606; Fax: +91 22 25505151; E-mail: vkaswal@ barc.gov.in.

’ REFERENCES (1) Suh, W. H.; Suh, Y. H.; Stucky, G. D. Nano Today 2009, 4, 27. (2) Leszczynski, J. Nat. Nanotechnol. 2010, 5, 633. (3) Rosen, M. J. Surfactants in Emerging Technologies; Marcel Dekker: New York, 1987. (4) Wasan, D. T.; Ginn, M.; Shah, D. O. Surfactants in Process Engineering; Marcel Dekker: New York, 1988. (5) Zhang, R.; Somasundaran, P. Adv. Colloid Interface Sci. 2006, 123126, 213. (6) You, C. C.; Miranda, O. R.; Gider, B; Ghosh, P. S.; Kim, I. B.; Erdogan, B.; Krovi, S. A.; Bunz, U. H. F.; Rotello, V. M. Nat. Nanotechnol. 2007, 2, 318. (7) Klein, J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 2029. (8) Lynch, I.; Dawson, K. A. Nano Today 2008, 3, 40. (9) Hu, M.; Qian, L.; Brinas, R. P.; Lymar, E. S.; Hainfeld, J. F. Angew. Chem., Int. Ed. 2007, 46, 5111. (10) Lynch, I.; Cedervall, T.; Lundqvist, M .; Lago, C. C.; Linse, S.; Dawson, K. A. Adv. Colloid Interface Sci. 2007, 134135, 167. (11) Shang, L.; Wang, Y.; Jiang, J.; Dong, S. Langmuir 2007, 23, 2714. (12) Berret, J.-F. Adv. Colloid Interface Sci. 2011 doi:10.1016/ j.cis.2011.01.008 (13) Rezwan, K.; Studart, A. R.; V€or€os, J.; Gauckler, L. J. J. Phys. Chem. B 2005, 109, 14469. (14) Qi, L.; Chapel, J.-P.; Castaing, J.-C.; Fresnais, J.; Berret, J.-F. Langmuir 2007, 23, 11996. (15) Wong, K.; Cabane, B.; Duplessix, R.; Somasundaran, P. Langmuir 1989, 5, 1346. (16) Golub, T P; Koopal, L K; Sidorova, M P Colloid J. 2004, 66, 38. (17) Qi, L.; Fresnais, J.; Berret, J.-F.; Castaing, J.-C.; Destremaut, F.; Salmon, J.-B.; Cousin, F.; Chapel, J.-P. J. Phys. Chem. C. 2010, 114, 16373. (18) Berret, J.-F.; Schonbeck, N.; Gazeau, F.; Kharrat, D. E.; Sandre, O.; Vacher, A.; Airiau, M. J. Am. Chem. Soc. 2006, 128, 1755. (19) Qi, L.; Sehgal, A.; Castaing, J.-C.; Chapel, J.-P.; Fresnais, J.; Berret, J.-F.; Cousin, F. ACS Nano 2008, 2, 879. (20) Berret, J. -F.; Yokota, K.; Morvan, M. Soft Mater. 2004, 2, 74. (21) Lugo, D. M.; Oberdisse, J.; Lapp, A.; Findenegg, G. H. J. Phys. Chem. B 2010, 114, 4183. (22) Alexeev, V. L.; Ilekti, P.; Persello, J.; Lambard, J.; Gulik, T.; Cabane, B. Langmuir 1996, 12, 2392. (23) Vertegel, A. A.; Siegel, R. W.; Dordick, J. S. Langmuir 2004, 20, 6800. (24) Rezwan, K.; Meier, L. P.; Gauckler, L. J. Biomaterials 2005, 26, 4351. (25) Rezwan, K.; Meier, L. P.; Rezwan., M.; V€or€os, J.; Textor, M.; Gauckler, L. J. Langmuir 2004, 20, 10055. (26) Lundqvist, M.; Sethson, I.; Jonsson, B. -H. Langmuir 2004, 20, 10639. (27) Turci, F.; Ghibaudi, E.; Colonna, M.; Boscolo, B.; Fenoglio, I.; Fubini, B. Langmuir 2010, 26, 8336. (28) Lundqvist, M.; Stigler, J.; Elia, G.; Lynch, I.; Cedervall, T.; Dawson, K. A. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 14265. (29) Rechendorff, K.; Hovgaard, M. B.; Foss, M.; Zhdanov, V. P.; Besenbacher, F. Langmuir 2006, 22, 10885.

ARTICLE

(30) Kane, R. S.; Stroock, A. D. Biotechnol. Prog. 2007, 23, 316. (31) De, M.; Miranda, O. R.; Rana, S.; Rotello, V. M. Chem. Commun. 2009, 2157. (32) Lacerda, D. P. S. H.; Park, J. J.; Meuse, C.; Pristinski, D.; Becker, M. L.; Karim, A.; Douglas, J. F. ACS Nano 2010, 4, 365. (33) Svergun, D. I.; Koch, M. H. J. Rep. Prog. Phys. 2003, 66, 1735. (34) Chodankar, S.; Aswal, V. K.; Kohlbrecher, J.; Vavrin, R.; Wagh, A. G. J. Phys.: Condens. Matter 2007, 19, 326102. (35) Aswal, V. K.; Kohlbrecher, J. Chem. Phys. Lett. 2006, 424, 91. (36) Kumar, S.; Aswal, V. K. J. Phys.: Condens. Matter 2011, 23, 035101. (37) Powroznik, B.; Gharbi, M.; Dandrifosse, G.; Peulen, O. Biochimie 2004, 86, 651. (38) Palacio, L.; Ho, C.-C.; Pradanos, P.; Hernandez, A.; Zydney, A. L. J. Membr. Sci. 2003, 222, 41. (39) Kohlbrecher, J.; Wagner, W. J. Appl. Crystallogr. 2000, 33, 804. (40) Hayter, J. B.; Penfold, J. Colloid Polym. Sci. 1983, 261, 1022. (41) Pedersen, J. S. Adv. Colloid Interface Sci. 1997, 70, 171. (42) Aswal, V. K.; Goyal, P. S.; Thiyagarajan, P. J. Phys. Chem. 1998, 102, 2469. (43) Teixeira, J. J. Appl. Crystallogr. 1988, 21, 781. (44) Pedersen, J. S. Adv. Colloid Interface Sci. 1997, 70, 171. (45) Pozzo, D. C.; Walker, L. M. Colloids Surf. A: Physicochem. Eng. Asp 2007, 294, 117. (46) Chodankar., S.; Aswal, V. K. Phys. Rev. E 2005, 72, 041931. (47) Shukla, A.; Mylonas, E.; Cola, E. D.; Finet, S.; Timmins, P.; Narayanan, T.; Svergun, D. I. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 5075. (48) Perevedentseva, E.; Cheng, C. -Y.; Chung, P. -H.; Tu, J. -S.; Hsieh, Y. -H.; Cheng, C. -L. Nanotechnology 2007, 18, 315102. (49) Chodankar, S.; Aswal, V. K.; Kohlbrecher, J.; Vavrin, R.; Wagh, A. G. Phys. Rev. E 2009, 79, 021912. (50) Roach, P.; Farrar, D.; Perry, C. C. J. Am. Chem. Soc. 2006, 128, 39395. (51) Wang, X.-Z.; Huang, Y. Phys. Rev. A 1992, 46, 5038. (52) Witten, T. A., Jr.; Sander, L. M. Phys. Rev. Lett. 1981, 47, 1400. (53) Jacrot, B. Rep. Prog. Phys. 1976, 39, 911. (54) M€uter, D.; Shin, T.; Deme, B.; Fratzl, P.; Paris, O.; Findenegg, G. H. J. Phys. Chem. Lett. 2010, 1, 1442.

10173

dx.doi.org/10.1021/la201291k |Langmuir 2011, 27, 10167–10173