Structure and Stability of Proteins Interacting with Nanoparticles - ACS

Chapter 38

Structure and Stability of Proteins Interacting with Nanoparticles Downloaded by UNIV OF ARIZONA on December 13, 2012 | http://pubs.acs.org Publication Date (Web): December 12, 2012 | doi: 10.1021/bk-2012-1120.ch038

Luigi Calzolai, Stefania Laera, Giacomo Ceccone, and Francois Rossi* European Commission, Joint Research Centre, Institute for Health and Consumer Protection Via E. Fermi 2749, I-21027, Ispra (VA), Italy *E-mail: [email protected]. Tel: +390332786561

The behavior and toxicological properties of nanoparticles (NP) in biological medium depends heavily on their interactions with proteins. In return, the structure, stability and biological properties of the proteins that interact with the nanoparticles are strongly affected by this interaction. Unfortunately, the mechanisms of interaction and their structural consequences are very difficult to analyse. Here we show the use of advanced biophysical techniques to obtain information on the structure and stability of protein-nanoparticle complexes. By using circular dichroism spectroscopy it is possible to detect changes in the secondary structure and stability of proteins upon interaction with nanoparticles. Moreover, by using nuclear magnetic resonance experiments, it is even possible to detect the specific domain of proteins interacting with nanoparticles.

Introduction Nanotechnology is having a large impact in very different scientific fields such as material sciences, photonics, nanomedicine and biotechnology. The uses of nanotechnology-based materials is not just limited to research laboratories, but has already been applied in several industrial sectors and into real products as disparate as medical diagnostic tools, drug delivery systems, cosmetics, and consumer products. In nanomedicine and nanotechnologies industries, the global market reached $63.8 billion in 2010 and $72.8 billion in 2011. The market is © 2012 American Chemical Society In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Downloaded by UNIV OF ARIZONA on December 13, 2012 | http://pubs.acs.org Publication Date (Web): December 12, 2012 | doi: 10.1021/bk-2012-1120.ch038

expected to grow to $130.9 billion by 2016 at a compound annual growth rate of 12.5% between years 2011 and 2016. (BCC Research Report, Nanotechnology in Medical Applications: The Global Market; April 2012). There is a growing public concern about the safety of engineered nanoparticles (ENPs) since it has been demonstrated that those intended for industrial and medical applications could cause adverse effects in mammalians or aquatic organisms by specific mechanisms depending on their physical chemical properties (1). Experiments have clearly shown that there are several factors that should be taken into account to understand the interaction of ENPs with organisms. While the physical chemical properties such as size distribution, composition, surface charge, and solubility influence cell viability, these effects are always mediated by biological entities adsorbed on the surface of the nanomaterial: in physiological environments, ENPs selectively absorb proteins to form NP-protein corona (2), a process governed by molecular interaction between chemical groups on the NP surface and the amino acids residues of the protein. When proteins interact with nanoparticles, their native conformation can be altered, thus reducing their stability or exposing new epitopes on the protein surface giving rise to unexpected biological responses that ultimately can lead to adverse effects (3, 4). In other cases, the binding of enzymes to nanoparticles can reduce (or increase) enzymatic activity then altering the normal cell homeostasis (5, 6). It is also becoming clear that the properties and fate of nanoparticles in biological systems, and ultimately the cellular responses to them, critically depends from the adsorbed biomolecule layer(s) (7, 8). The identity of the proteins adsorbed on the nanoparticle surface will change the interface properties and strongly influence the interaction of the nanoparticles with cells, thus modifying both the cellular uptake and distribution (9). In this chapter, we will focus on what happens to nanoparticles when entering biological systems and some of the available techniques to study in detail this complex problem. Finally we will show recent results from our experimental work on the use of biophysical techniques to analyze the changes in structure and stability of proteins in protein-nanoparticle complexes.

What Happens To NP in Biological Systems? When nanoparticles enter into contact with biological systems (cells, tissues, or biological fluids) their size, stability, electrostatic potential, and surface chemical properties can change significantly. In fact, biological systems contain several components that can greatly influence the physico-chemical properties of nanoparticles and thus their interaction with living systems. Even simple biological fluids such as serum or complete cell culture medium contain various salts (for example NaCl, phosphates) in quite high concentrations, a multitude of different proteins, ligands (such as chloride ions, citrate, sulphides), reducing agents (such as glutathione) as well as oxidizing agents. Figure 1 schematically depicts the possible behaviour of NP in a typical in-vitro cell system. A NP (black 840 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


spheres) of a well defined and monodispersed size can aggregate (red circle) due to the high ionic strength of the biological medium. Single Nanoparticles, or NP aggregates, can then bind to proteins present in the medium (yellow circle) and this NP-protein complex can enter inside cells. Inside the cell, the NP-protein complex will interact with different cell compartments (such as mitochondria, lysosome, and nucleus), possibly causing damages that ultimately lead to toxicity. This broad and general picture is even more complex in the case of some metal and metal oxide nanoparticles where the oxidative release of metal ions seems to be the most significant contributor to the nanotoxicity, especially in the case of silver nanoparticles (10, 11). For instance, the metallic atom present in silver nanoparticles can be oxidized to silver ions by dissolved oxygen molecules; in biological systems are present several compounds (such as cysteine, chloride, thiols, citrate) that can contribute to modulate the ion release rate over 4 orders of magnitude (12). This mechanism of cell nanotoxicity seems not to be limited to silver nanoparticles, but it has been recently shown to be active also in the case of zinc oxide and copper oxide (13).

Figure 1. Nanoparticles in biological systems.

How To Study Protein−Nanoparticle Interactions The interaction of nanomaterials with proteins is a difficult problem to study due both to the complexity of the system involving a solid/liquid interface (14) and low concentration of the proteins to be analyzed. The characterization of nanomaterials in biological systems (even relatively simple ones such as serum) is much more challenging due to the complexity and heterogeneity of biological medium per se. 841 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


The protein-nanoparticle interactions can be approached using a multistep process starting from the basic problem of measuring the changes in the size of nanoparticles, quantifying the amount of protein(s) adsorbed onto them, and to the more sophisticated identification of the proteins forming the protein corona, to measuring the structure and stability changes to the protein forming the NP-protein complexes. To measure the size of NP, there are several techniques available, but none of them is completely suitable for measuring changes in the size of NP interacting with proteins in biological systems. Generally speaking, electron microscopy (either SEM or TEM) is one of the best methods for measuring the size of particles (especially in the case of NP containing heavy atoms) and is one of few that can address the problem of measuring sizes of NP agglomerates. Unfortunately, even this method is not very well suited to evaluate protein adsorption due to the requirements for sample preparation and poor contrast given by protein molecules. In fact, EM instruments work under high vacuum and proteins are mainly composed of atoms of low atomic number which generally do not provide enough contrast in the EM micrographs to obtain a clear image of the protein corona surrounding the metallic NP core. To measure the size of NP in solutions, one of the most used techniques is dynamic light scattering, DLS, (sometimes referred as photon correlation spectroscopy). The technique measures the autocorrelation function of particles tumbling in solution and then by a Laplace-type transformation calculates the particle size distribution. The technique works quite well for NP made up by a single size but is not very reliable in the case of multiple sizes due to the fact that the intensity of signal of the various species is proportional to the 6th power of the size and thus the presence of even very small amounts of large aggregates would cover the signal from smaller particles (15). The technique is suitable for a careful use in well defined conditions with limited complexity. For example, Figure 2a shows the DLS data for free gold nanoparticles (AuNP) of around 20nm and then of the AuNP in complex with human serum albumin (HSA) of around 30nm that nicely fit with the presence of a single monolayer of albumin protein (HSA has a size of around 5nm) around the gold nanoparticles. It must be noted that such results can be obtained with DLS in carefully optimized experimental conditions (such as neutral pH and very low ionic strength) that minimizes the amount of aggregates. In more complex (and potentially more relevant) cases such as the analysis of NP in serum medium, other techniques show a great potential. For example our group (15) and others (16) have found that the combination of a size separation technique such as flow field flow fractionation and size measurement with light scattering detectors (DLS or MALS) offers a very powerful combination for the characterization of the samples. Figure 2b shows an example of the use of FFF coupled with a light scattering detector to characterize SiO2NP-protein corona. On the horizontal axis, the exit time from the separation channel is proportional to the size of the particles, with smaller particles exiting first, on the vertical axis the signal of the light scattering detector at 90° is related to the amount of material and to the size. The blue curve shows the SiO2NP (size 50nm) in water, exiting at 28′, the orange curve is the plasma serum alone showing some small proteins exiting at 842 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


8′ and larger proteins at 18′. The red curve is the SiO2NP sample in plasma serum: it shows peaks at around 9′ and 18′ (due to the free plasma proteins) and a peak at 34′ due to the SiO2NP-protein corona complex.

Figure 2. Measurement of the size of nanoparticle-protein complexes. (a) Dynamic light scattering measurement of free AuNP (red) and in complex with human serum albumin (green). (b) Field flow fractionation separation of SiO2 nanoparticles in plasma serum with light scattering detector. Plasma serum alone (orange); free SiO2NP (blue), SiO2NP in plasma serum (red).

After separating the NP-protein corona system, it is possible to detect the proteins forming the corona by using proteomics approaches that identify each single protein using mass spectroscopy-based techniques (17–19). The results of such studies show that the protein corona surrounding nanoparticles is a dynamic system with a complex time evolution. In general the most abundant proteins in plasma serum do bind first to nanoparticles, while other less abundant proteins with 843 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


stronger interaction with the NP replace them at later times (20, 21). This effect reminds to the so called Vroman effect, that describes how blood serum proteins absorb to a surface: the highest mobility proteins generally arrive first and are later replaced by less motile proteins that have a higher affinity for the surface (22, 23). The studies performed so far have focused on the identification of the proteins forming the protein corona at different time points, and on the evaluation of the kinetic constants of the interaction of the different proteins with the nanoparticles (2, 9, 20, 24). Various techniques have been used for measuring the affinity of proteins for different nanoparticles, such as isothermal titration calorimetry (25) and surface plasmon resonance (2). In addition, size exclusion chromatography has been used to determine the residence times of proteins (starting from a complex protein mixture) on nanoparticles (2). The effect of nanoparticle binding on the three dimensional structure and stability of the proteins is clearly an important issue, but up to now has received less attention, probably due to the inherent difficulties of the experimental system to be investigated.

How To Study the Structure of Proteins in NP−Protein Systems The analysis of the structure, stability and function of proteins in protein−NP complexes is a key requirement to understand if and how the properties of the various proteins change when bound/interacting with NP. In principle, the structural analysis of proteins in the complex formed with nanoparticles can be performed with the available techniques normally used in structural biology: in a sense, studying protein−NP interaction is not much different than studying protein-protein interaction. Unfortunately, in the case of NP−proteins complexes, there are major complications: the system can be dynamic, for example the protein corona is a very dynamic and heterogeneous system where the amounts and even the identities of the involved proteins changes over time. This thus rules out the use of crystals and X-ray spectroscopy that is somehow the workhorse of structural biology. On the other hand, nuclear magnetic resonance is well adapted to studying dynamic systems and later on, an example of the use of NMR in identifying the protein-gold nanoparticle interaction site will be provided. Among the methods that can give low resolution structural information, circular dichroisms (CD) has several advantages: it can rapidly evaluate the secondary structure, folding, and binding properties of proteins (26, 27) and, critically, the technique is very sensitive and CD spectra (under optimal conditions) can be acquired with just a few micrograms of sample. The 180-250 nm region of the circular dichroism spectrum of a protein is sensitive to the secondary structure elements present in the protein. Alpha-helical regions have a typical spectrum with a double minimum at 222 and 208 nm and a maximum at 190nm, while beta-sheet structures have a single minimum at 215 nm and a maximum at 190nm. Using a basis set of CD spectra of secondary structures, it is possible to deconvolute the CD spectrum of a given protein and thus estimate the amount of each secondary structure elements present. Several open access programs are available to perform such deconvolution, and some of them (such as 844 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


the dichroweb tool) are accessible via a web browser interface, thus providing an easy and simple access to such tools even to non specialists. With “standard” CD, that uses a normal lamp as irradiation source, it is possible to detect the secondary structure of proteins and their changes upon interactions with nanoparticles. For example the black spectrum in Figure 3 shows the CD spectrum of human serum albumin (HSA). Using a 1cm long quartz cell, it is possible to measure the CD spectra of HSA with a low concentration of 5ug/mL (corresponding to a molar concentration of 100 nM). The red spectrum in Figure 3 shows the CD spectrum of HSA in the presence of silicon dioxide nanoparticles (SiO2-NP) of 50nm size. From the two spectra, it is clear that upon interaction with SiO2-NP, the HSA protein changes its secondary structure. Using the dichroweb program to estimate the amount of secondary structure elements present in the protein, it is found that when HSA interacts with SiO2NP, the amount of alpha-helix decreases by around 12%, while the β-sheet and turn content increase by 8% and 3%, respectively. One of the problem of measuring CD spectra in the far UV region at the lowest possible concentrations, is represented by the absorbance of buffers and salts present in the sample that do not allow the collection of meaningful data below the 190 nm wavelength. This causes even more problems when measuring protein−NP complexes, where the NP (especially silver, gold, and silicon dioxide) show strong absorbance below 190nm. This effect could be minimized by reducing the cell pathlenght, but in this case higher protein concentrations should be used, which would change the protein NP ratio and produce agglomeration.

Figure 3. CD spectra acquired with standard bench top instrument. CD spectrum of HSA (black) and HSA-SiO2NP (red).

To be able to measure CD spectra of even less concentrated proteins, long path measuring cells should be used, but to accomplish this there is the need of high flux and highly collimated excitation sources. Synchrotron radiation (SR) has these characteristics and the use of such sources to perform CD experiments (SRCD) presents several advantages compared to conventional CD techniques. 845 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


Using the B23 beam line of Diamond Light Source, we have recently shown that it is possible to collect high quality CD spectra of proteins with just 1 μg/mL samples (28). Figure 4 shows the CD spectra of HSA and lysozyme collected using a cell with a 10cm long pathway and 800 μL volume sample. The spectrum of HSA has been collected with a concentration of 20nM, and given its quality, it would be possible to collect CD spectra with concentration as low as 8nM with still a good signal to noise ratio by increasing the number of scans from 4 to 16.

Figure 4. SRCD spectra of low concentration proteins acquired with a 10cm pathlenght cell. Insets: cartoon picture of the protein secondary structures with alpha-helices in red and beta-sheets in cyan. (a) Human serum albumin. (b) Human Lysozyme. Reproduced with permission from ref. (28). Copyright 2011 American Chemical Society. The possibility of working at such a low protein concentration with SRCD allows measuring the changes in secondary structure associated with proteins bound noncovalentely to metallic nanoparticles with a protein-particle ratio in the range corresponding to a monolayer. Figure 5a shows the CD spectra of free HSA (black spectrum) and in complex with silver nanoparticles (red spectrum) in a molar ratio of 44:1 and that of lysozyme free (black) and in complex with silver nanoparticles (red) in a molar ratio of 200:1. The CD spectrum of HSA-AgNP shows that human serum albumin does not change its overall secondary structure upon interaction with silver nanoparticles. On the contrary, the CD spectrum of lysozyme-AgNP complex in Figure 5b clearly indicates a significant change in the intensity and shape. The data suggests that part of the lysozyme protein precipitates upon interaction with AgNP. This is confirmed also by the shift towards bluish colour of silver nanoparticles (indicative of the formation of particles larger than 80-100nm) and by DLS measurements that indicate the formation of large aggregates in the lysozyme-AgNP sample. 846 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


This different behaviour between serum albumin and lysozyme upon interaction with silver nanoparticles can be explained by the different electrostatic surface potential properties of the two proteins. Serum albumin has an isoelectric point of 5.5 and thus in the experimental conditions of pH 7.0, its charge is overall negative (with some positive patches, see inset in Figure 5a). On the contrary lysozyme has an isoelectric point of 9.5 and at pH 7.0 it is almost completely positively charged (inset in Figure 5b), thus greatly enhancing its interaction with the negatively charged silver nanoparticles, that will cause the nanoparticles to loose their repulsive forces leading to aggregation and partial precipitation of the system.

Figure 5. Changes in protein structure due to protein-silver nanoparticles interaction. (a) CD spectrum of free HSA (black) and HSA-AgNP (red). (b) CD spectra of lysozyme (black) and lysozyme-AgNP (red). Reproduced with permission from ref. (28). Copyright 2011 American Chemical Society.

Measuring Stability Changes of Protein−NP Systems Circular dichroism can also be used to assess the thermal stability of proteins, and, in the case of reversible thermal unfolding, it is possible to calculate the Gibbs free energy of the unfolding process. The technique is quite sensitive and can be used to assess subtle differences in the stability of proteins in different conditions, for example the destabilization of human prion proteins in acidic conditions, as compared to neutral pH (29). Using SRCD, we have studied the thermal unfolding process of serum albumin in the presence of different metallic nanoparticles. The whole CD spectrum of the protein has been collected varying the temperature from 20° C to 90° C at intervals of 2° C; a subset of the CD spectra at variable temperatures is shown in Figure 6a. This ensemble of spectra has been analyzed by singular value decomposition to identify the singular vectors that have the highest singular values (30). This analysis indicated that the whole dataset could be described by the combination of just two singular vectors, while attempt to involve a third vector resulted in it giving a negligible contribution. The CD spectra corresponding to the 847 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


two components identified by the singular value decomposition of the thermal unfolding process of free HSA are shown in Figure 6b: component 1 (black spectrum) corresponds to the completely folded spectrum of HSA, while the component 2 (red spectrum) is clearly the CD spectrum of a random coil protein devoided of any secondary structure. This indicates that the thermal unfolding process of HSA can be described as a two-states process going from a fully folded to a fully unfolded structure. Each one of the CD spectra acquired at variable temperatures is thus a linear combination of the two components with different weightings of the two basis spectra. Plotting the percentage of component 1 presents in each CD spectrum as a function of temperature gives a typical sigmoid curve shown in Figure 6c.

Figure 6. Measuring the thermal unfolding of HSA at 5ug/mL by SRCD. (a) CD spectra collected at variable temperatures from 20 C to 90 C. (b) CD spectra of the two components identified by singular value decomposition. (c) Non-linear square fitting (red curve) of % of folded structure as a function of temperature (black squares). Reproduced with permission from ref. (28). Copyright 2011 American Chemical Society. Using the same approach we have then measured the thermal unfolding process of HSA in the presence of gold nanoparticles (AuNP) and silver nanoparticles (AgNP). Due to the use of SRCD we have been able to measure high quality data using quite low concentrations of proteins, and a molecular ratio of 22:1 for the HSA-AuNP system and 18:1 for the HSA-AgNP system. Figure 7 shows the thermal unfolding data for the three systems. The melting temperature (TM) of HSA in complex with gold nanoparticles is 74.8±1.3°C, thus experimentally indistinguishable from the TM of the protein alone, 75.2±0.6°C. In contrast, the TM of the HSA-AgNP system is 69.1±1.0°C, significantly lower than that of the protein alone or in complex with gold nanoparticles. The shape of the three sigmoid thermal unfolding curves indicate that in the case of the HSA-AgNP complex the unfolding transition is less steep and less cooperative compared to the free protein. The thermal unfolding process of serum albumin is not reversible and thus it is not possible to extract real thermodynamic parameters from the data, but it is possible to use the values of the melting temperatures to evaluate the relative stability of HSA interacting with gold and silver nanoparticles. Analyzing the 848 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


CD spectra and the thermal unfolding of the protein in the three conditions (free, +AuNP, +AgNP) it can be concluded that upon interaction with silver nanoparticles, the protein significantly reduces its thermal stability and seems to adopt a less compact folded state. On the contrary, the interaction with gold nanoparticles does not alter the stability of the protein and its overall folded state. In summary, the use of SRCD allows measuring the changes in the secondary structure of proteins at unprecedented low concentration (as low as a few nanomolar concentration) and provides a unique tool for characterizing the structural properties of proteins at bio-nonbio interfaces. In addition, the analysis of CD-detected thermal unfolding allows to measure changes in the relative stability of proteins interacting with different nanomaterials and gives access to information that are difficult to obtain with other techniques, especially at these very low concentrations.

Figure 7. Destabilization of HSA-AgNP system. SRCD detected HSA thermal unfolding. Plot of % of folded protein as a function of temperature (black squares), fitted to Boltzmann equation (red curve). (a) free HSA protein. (b) HSA-AuNP. (c) HSA-AgNP. Reproduced with permission from ref. (28). Copyright 2011 American Chemical Society.

High-Resolution Structure of Protein−NP Complexes High resolution informations on the structure of proteins bound to nanoparticles are very sparse and very difficult to obtain due to the challenges imposed by system involving the bio/non-bio interfaces. The protein−NP system is, in general, quite dynamic, and this makes the use of X-ray diffraction almost impossible, as usually dynamic systems do not crystallize and well behaving crystals are a prerequisite for X-ray protein crystallography. Nuclear Magnetic Resonance (NMR) is a well established technique to obtain high resolution three dimensional structures of proteins in solution (31), and is particularly well suited to obtain information on weak protein-protein interactions (32, 33). NMR has been used to detect the interaction of nanoparticles with proteins by using either hydrogen-deuterium exchange experiments or two-dimensional NMR 849 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


spectroscopy (17, 34). By using 2D NMR spectroscopy and chemical shift perturbation we have shown that it is possible to characterize at the amino acid level the interaction between the ubiquitin protein and gold nanoparticles (35). Our approach relied on the measurement of two dimensional NMR spectra that are able to detect the 1H and 15N chemical shifts (technically 15N-1H heteronuclear single quantum correlation, [15N-1H]-HSQC) of each N-H group present in a 15N-labelled protein. A typical HSQC spectrum for human ubiquitin is shown in Figure 8 with highlighted one amide N-H group of the backbone of the protein that gives rise to a single 2D peak on the two-dimensional spectrum. With modern NMR instruments equipped with cryoprobe it is possible to measure 2D NMR experiments with protein samples with concentrations of the order of 10-20 μM, that for NMR experiments are remarkably low.

Figure 8. Two dimensional (15N-1H)-HSQC spectrum of 15N-labelled human Ubiquitin in complex with gold nanoparticles. Each peak on the spectrum represents one N-H chemical group of the protein, mainly from the amide backbone (left inset). Reproduced with permission from ref. (35). Copyright 2010 American Chemical Society.

With some effort and the acquisition of several NMR experiments each one of the 2D peaks present in the HSQC spectrum can be sequence-specifically assigned to each amino acid present in the protein. The position (1H and 15N chemical shifts) of each peak is very sensitive to the chemical environment and somehow represents the environment “sensed” by each amino acid in the protein. By collecting 2D HSQC NMR experiments of two samples, one of the free protein and one of the protein in the presence of nanoparticles, it is thus possible to map the “chemical perturbation” induced by nanoparticles on the protein side at the amino acid level. These effects can be monitored by quantifying the chemical shift changes in terms of the chemical shift perturbation (CSP), defined as (33):

850 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


where Δ1H is the difference in chemical shift on the 1H dimension between free protein and bound protein for each amino acid and Δ15N is the corresponding chemical shift difference for the 15N dimension. Figure 9a shows the CSP for the interaction of ubiquitin with gold nanoparticles of 10nm size, mapped onto the amino acid sequence of the protein. The data shows that the vast majority of the amino acids show changes smaller than 0.02 ppm that are almost negligible. Two peptide fragments (Gln2-Ile3, and Leu15-Glu18) show substantial changes. The fact that the biggest changes are not isolated, but cluster in groups of more than 1 amino acid gives confidence that the detected changes are not artefacts due for example to incorrect assignment of the NMR peaks.

Figure 9. Chemical shift perturbation data of interaction of ubiquitin with gold nanoparticles. (a) CSP mapped onto the amino acid sequence of the protein. (b) CSP mapped onto the three dimensional cartoon structure of the protein. Reproduced with permission from ref. (35). Copyright 2010 American Chemical Society. 851 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


These results become much more interesting when mapped onto the well known three dimensional structure of the protein. Figure 9b shows the cartoon structure of the backbone of human ubiquitin, with the two peptide fragments showing the largest CSP changes coloured in red. The two fragments formed by the amino acids 2-3 and 15-18 are apart in the amino acid sequence, but are actually close in space due to the globular folding of the protein. They partially belong to the beta strand 1 and beta strand 2 that form the first β-sheet of ubiquitin. The fragments Gln2-Ile3 and Leu15-Glu18, identified by CSP using NMR, are a well defined part of the ubiquitin protein that interacts with gold nanoparticles, thus forming a gold-nanoparticle interaction domain. A similar experimental approach can be applied to other protein-nanoparticle systems to identify specific interacting or binding domains on the protein side. The main requirements for such studies is the use of modern NMR instruments equipped with cryoprobes and the availability of recombinant proteins labelled with 15N (and eventually 13C).The use of cryoprobes enhances the signal to noise ratio of NMR experiments of a factor 8 to 10 and thus allows to collect 2D spectra of protein samples with concentration as low as 10-20 μM and sample volumes of a minimum of 300μL (with the use of specialized NMR tubes). One of the main drawbacks of NMR is posed by the maximum size of the proteins and of the protein-nanoparticle complex that can be reasonably measured. Recording NMR data of proteins of up to 50KDa in size (corresponding to around 500 amino acids) is quite standard nowadays (36) and it is also possible to measure 2D experiments of protein-antibody complexes of around 200 KDa (37). Our results show that the interaction of a small protein such as ubiquitin with gold nanoparticles is quite specific and that by using NMR it is possible to obtain high resolution structural information, at the amino acid level of detail, on protein-nanoparticle complexes. The availability of such detailed experimental information will allow a better understanding of the structural signatures that guide the interaction of proteins with nanoparticles and will also open up new avenues of research in the detailed modelling and eventual prediction of such interactions.

References 1.

2.

3. 4.

Oberdorster, G.; Oberdorster, E.; Oberdorster, J. Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles. Environ. Health Perspect. 2005, 113 (7), 823–39. Cedervall, T.; Lynch, I.; Lindman, S.; Berggard, T.; Thulin, E.; Nilsson, H.; Dawson, K. A.; Linse, S. Understanding the nanoparticle-protein corona using methods to quantify exchange rates and affinities of proteins for nanoparticles. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (7), 2050–5. Lynch, I.; Dawson, K. A.; Linse, S. Detecting cryptic epitopes created by nanoparticles. Sci. STKE 2006, 2006 (327), pe14. Deng, Z. J.; Liang, M.; Monteiro, M.; Toth, I.; Minchin, R. F. Nanoparticle-induced unfolding of fibrinogen promotes Mac-1 receptor activation and inflammation. Nat. Nanotechnol. 2011, 6 (1), 39–44. 852 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

5. 6.

7. 8.


9.

10.

11. 12.

13.

14.

15.

16.

17.

18.

19.

Wu, Z.; Zhang, B.; Yan, B. Regulation of Enzyme Activity through Interactions with Nanoparticles. Int. J. Mol. Sci. 2009, 10 (10), 4198–4209. Speshock, J.; Braydich-Stolle, L.; Szymanski, E.; Hussain, S. Silver and Gold Nanoparticles Alter Cathepsin Activity In vitro. Nanoscale Res. Lett. 2011, 6 (1), 17. Gray, J. J. The interaction of proteins with solid surfaces. Curr. Opin. Struct. Biol. 2004, 14 (1), 110–5. Norde, W.; Lyklema, J. Why proteins prefer interfaces. J. Biomater. Sci.: Polym. Ed. 1991, 2 (3), 183–202. Walczyk, D.; Bombelli, F. B.; Monopoli, M. P.; Lynch, I.; Dawson, K. A. What the cell “sees” in bionanoscience. J. Am. Chem. Soc. 2010, 132 (16), 5761–8. Lok, C. N.; Ho, C. M.; Chen, R.; He, Q. Y.; Yu, W. Y.; Sun, H.; Tam, P. K.; Chiu, J. F.; Che, C. M. Silver nanoparticles: partial oxidation and antibacterial activities. J. Biol. Inorg. Chem. 2007, 12 (4), 527–34. Liu, J.; Hurt, R. H. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environ. Sci. Technol. 2010, 44 (6), 2169–75. Liu, J.; Sonshine, D. A.; Shervani, S.; Hurt, R. H. Controlled release of biologically active silver from nanosilver surfaces. ACS Nano 2010, 4 (11), 6903–13. Zhang, H.; Ji, Z.; Xia, T.; Meng, H.; Low-Kam, C.; Liu, R.; Pokhrel, S.; Lin, S.; Wang, X.; Liao, Y. P.; Wang, M.; Li, L.; Rallo, R.; Damoiseaux, R.; Telesca, D.; Madler, L.; Cohen, Y.; Zink, J. I.; Nel, A. E. Use of Metal Oxide Nanoparticle Band Gap To Develop a Predictive Paradigm for Oxidative Stress and Acute Pulmonary Inflammation. ACS Nano 2012, 6 (5), 4349–4368. Nel, A. E.; Madler, L.; Velegol, D.; Xia, T.; Hoek, E. M.; Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M. Understanding biophysicochemical interactions at the nano-bio interface. Nat. Mater. 2009, 8 (7), 543–57. Calzolai, L.; Gilliland, D.; Garcia, C. P.; Rossi, F. Separation and characterization of gold nanoparticle mixtures by flow-field-flow fractionation. J. Chromatogr., A 2011, 27, 1218. Schmidt, B.; Loeschner, K.; Hadrup, N.; Mortensen, A.; Sloth, J. J.; Koch, C. B.; Larsen, E. H. Quantitative characterization of gold nanoparticles by field-flow fractionation coupled online with light scattering detection and inductively coupled plasma mass spectrometry. Anal. Chem. 2011, 83 (7), 2461–8. Lundqvist, M.; Stigler, J.; Elia, G.; Lynch, I.; Cedervall, T.; Dawson, K. A. Nanoparticle size and surface properties determine the protein corona with possible implications for biological impacts. Proc. Natl. Acad. Sci. U.S.A. 2008, 105 (38), 14265–70. Lundqvist, M.; Stigler, J.; Cedervall, T.; BerggÃ¥rd, T.; Flanagan, M. B.; Lynch, I.; Elia, G.; Dawson, K. The Evolution of the Protein Corona around Nanoparticles: A Test Study. ACS Nano 2011, 5 (9), 7503–7509. Casals, E.; Pfaller, T.; Duschl, A.; Oostingh, G. J.; Puntes, V. Time Evolution of the Nanoparticle Protein Corona. ACS Nano 2010, 4 (7), 3623–3632. 853 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


20. Cedervall, T.; Lynch, I.; Foy, M.; Berggard, T.; Donnelly, S. C.; Cagney, G.; Linse, S.; Dawson, K. A. Detailed identification of plasma proteins adsorbed on copolymer nanoparticles. Angew. Chem., Int. Ed. 2007, 46 (30), 5754–6. 21. Dobrovolskaia, M. A.; Patri, A. K.; Zheng, J.; Clogston, J. D.; Ayub, N.; Aggarwal, P.; Neun, B. W.; Hall, J. B.; McNeil, S. E. Interaction of colloidal gold nanoparticles with human blood: effects on particle size and analysis of plasma protein binding profiles. Nanomedicine 2009, 5 (2), 106–17. 22. Vroman, L. Effect of absorbed proteins on the wettability of hydrophilic and hydrophobic solids. Nature 1962, 196, 476–7. 23. Vroman, L.; Adams, A. L.; Fischer, G. C.; Munoz, P. C. Interaction of high molecular weight kininogen, factor XII, and fibrinogen in plasma at interfaces. Blood 1980, 55 (1), 156–9. 24. Monopoli, M. P.; Walczyk, D.; Campbell, A.; Elia, G.; Lynch, I.; Bombelli, F. B.; Dawson, K. A. Physical-chemical aspects of protein corona: relevance to in vitro and in vivo biological impacts of nanoparticles. J. Am. Chem. Soc. 2011, 133 (8), 2525–34. 25. Lindman, S.; Lynch, I.; Thulin, E.; Nilsson, H.; Dawson, K. A.; Linse, S. Systematic investigation of the thermodynamics of HSA adsorption to N-isopropylacrylamide/N-tert-butylacrylamide copolymer nanoparticles. Effects of particle size and hydrophobicity. Nano Lett. 2007, 7 (4), 914–20. 26. Fasman, G. D. Circular dichroism and the conformational analysis of biomolecules; Plenum Press: New York, 1996; p ix, 738 p. 27. Greenfield, N. J. Using circular dichroism spectra to estimate protein secondary structure. Nat. Protoc. 2006, 1 (6), 2876–90. 28. Laera, S.; Ceccone, G.; Rossi, F.; Gilliland, D.; Hussain, R.; Siligardi, G.; Calzolai, L. Measuring protein structure and stability of protein-nanoparticle systems with synchrotron radiation circular dichroism. Nano Lett. 2011, 11 (10), 4480–4. 29. Calzolai, L.; Zahn, R. Influence of pH on NMR structure and stability of the human prion protein globular domain. J. Biol. Chem. 2003, 278 (37), 35592–6. 30. Konno, T. Conformational diversity of acid-denatured cytochrome c studied by a matrix analysis of far-UV CD spectra. Protein Sci. 1998, 7 (4), 975–82. 31. Wüthrich, K. NMR of proteins and nucleic acids; Wiley: New York, 1986; p xv, 292. 32. Vaynberg, J.; Qin, J. Weak protein-protein interactions as probed by NMR spectroscopy. Trends Biotechnol. 2006, 24 (1), 22–7. 33. Zuiderweg, E. R. Mapping protein-protein interactions in solution by NMR spectroscopy. Biochemistry 2002, 41 (1), 1–7. 34. Engel, M. F.; Visser, A. J.; van Mierlo, C. P. Conformation and orientation of a protein folding intermediate trapped by adsorption. Proc. Natl. Acad. Sci. U.S.A. 2004, 101 (31), 11316–21. 35. Calzolai, L.; Franchini, F.; Gilliland, D.; Rossi, F. Protein--nanoparticle interaction: identification of the ubiquitin--gold nanoparticle interaction site. Nano Lett. 2010, 10 (8), 3101–5. 36. Horst, R.; Bertelsen, E. B.; Fiaux, J.; Wider, G.; Horwich, A. L.; WÃ¼thrich, K. Direct NMR observation of a substrate protein bound to 854 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.


the chaperonin GroEL. Proc. Natl. Acad. Sci. U.S.A. 2005, 102 (36), 12748–12753. 37. Simonelli, L.; Beltramello, M.; Yudina, Z.; Macagno, A.; Calzolai, L.; Varani, L. Rapid structural characterization of human antibody-antigen complexes through experimentally validated computational docking. J. Mol. Biol. 2010, 396 (5), 1491–507.

855 In Proteins at Interfaces III State of the Art 2012; Horbett, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2012.

Structure and Stability of Proteins Interacting with Nanoparticles - ACS

Recommend Documents