- Email: [email protected]
Gold nanoparticle-based tool to study protein conformational variants: implications in hemoglobinopathy
Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14 – 19 www.nanomedjournal.com
Clinical Nanomedicine
Gold nanoparticle-based tool to study protein conformational variants: implications in hemoglobinopathy Jaydeep Bhattacharya, MTech, Sinu Jasrapuria, MSc, Tapan Sarkar, MSc, Ranjita GhoshMoulick, MSc, Anjan Kr. Dasgupta, PhD4 Department of Biochemistry, Calcutta University, Calcutta, India Received 13 April 2006; accepted 24 October 2006
Abstract
The size of gold nanoparticles is shown here to gradually decrease if it is allowed to grow on a protein template, and the protein is subjected to unfolding by a nonionic denaturant. The correlation between size of the gold nanoparticle formed and the plasmon frequency observed remains linear, except at stages where protein folding intermediates are formed. Higher population of exposed tyrosine residues, number of sulfhydryl groups of the protein, and the overall exposition of the inner hydrophobic core may lead to the generation of smaller particles. The method provides a simple colorimetric sensing of protein conformation and has been tested for both nonheme and heme proteins (hemoglobin and bovine serum albumin). Similarly, protein variants with defects in folding (caused by subunit misassembly or mutation) can also be classified. Possible application of this approach in hemoglobinopathy (e.g., thalassemia carrier detection) is discussed in the text. D 2007 Published by Elsevier Inc.
Key words:
Gold nanoparticle; Plasmon resonance; Folding; Hemoglobinopathy; Hemoglobin; Serum albumin
Gold nanoparticles (GNPs) have been recently used in sensing folding states of proteins [1]. The observation has a wide range of implications in protein science, because it provides a simple colorimetric tool to follow the thermodynamic transitions occurring at different scales. The extent of reduction and the geometry of the reducing surface are known to play important roles in dictating the size of the emergent nanoparticles. The mechanism of the sensing technique relies on the size of the GNP, which in turn depends on the immediate reducing environment. A standard method of preparation of GNPs is sodium borohydride- or citrate-mediated reduction of aurochloric acid (HAuCl4). The particles thus produced should have a stabilized geometrical configuration to be
No conflict of interest was reported by the authors of this paper. 4 Corresponding author. Calcutta University, Calcutta, India. E-mail address: [email protected] (A.K. Dasgupta). 1549-9634/$ – see front matter D 2007 Published by Elsevier Inc. doi:10.1016/j.nano.2006.10.159
useful as sensors. If a protein template is present the reducing environment is altered. Amino groups or exposed thiol groups [2,3] present on the peptide surface can offer an environment that may facilitate emergence of size-stabilized particles. Again, exposed tyrosines may aid the reduction process, where an electron from the tyrosinate ion of the peptide is transferred to the metal ion at basic pH through the formation of a tyrosyl radical. The latter is converted to its dityrosine form during the reaction [4]. Recently, Zhou et al. prepared gold-silk fibroin core-shell nanoparticles by using tyrosine residues as both reducing and stabilizing agents, but they did not report the detailed nanoparticle formation mechanism [5]. Sastry and colleagues reported silver ion reduction with tyrosine at high pH [6]. In the investigation discussed here, a controlled alteration of the reducing surface was enforced by denaturants. Partial exposure of amino acid residues as a result of unfolding led to emergence of nanoparticles with a graded size distribution. The technique becomes particularly useful in sensing
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
15
16
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
the folding states of protein under identical buffering condition in response to denaturing stress. It may be important to cite Zare and co-workersT demonstration [1] once again. Because the unfolding was initiated by pH change the redox state of the reaction mixture might alter. It is tempting to ask whether the different colors are true representatives of different folding states or representatives of differential chemical activity of reducing agents originating in altered pH environments. We chose the nonionic denaturant urea to unfold the protein, so that the redox environment of the reaction mixture would not be disturbed. Eventually we could obtain a condition under which the unfolding studied using alternative methods and reported earlier [7] could be sensed by the plasmon frequency changes (as also reflected by size change of the GNPs). Methods Protein sample preparation Blood samples were collected in ethylenediamine tetraacetate (EDTA) vials from normal healthy persons. The whole blood was centrifuged at 3,000 g for 10 minutes and washed three times in normal saline. Hemolysis was carried out by mixing 100 AL of packed erythrocytes in 500 AL of lysis buffer (2.5 M NaCl, 100 mM EDTA-Na salt, 1% Triton X-100, and 10% dimethyl sulfoxide) [8]. The solution was applied to a Sephadex G-50 column equilibrated with 0.05 M phosphate buffer pH 7.0. Elution was carried out with 0.05 M phosphate buffer pH 7.0 at a flow rate of 25 mL/h. BBSA (10 mg/mL) was dissolved in 0.1 M phosphate buffer. Preparation of protein-free GNPs GNPs were prepared by the reduction of HAuCl4) by sodium borohydride. The normal reduction process was carried out according to the standard protocol. The size of GNPs obtained by this process was 14 nm [1]. Preparation of GNPs using proteins as templates The combined reducing property of sodium borohydride and the tyrosine of the protein HbA0 and BSA were used to reduce HAuCl4 to form GNPs. The major change that was made in this process was that the concentration of the sodium borohydride was kept low enough that it could not alone reduce HAuCl4 to form GNPs; yet the combined effect of the two reducing agents was capable of producing GNPs.
Fig 2. The mean hydrodynamic diameter (dh) of GNPs is plotted against urea concentration. The arrow indicates position where the folding intermediates may be formed. The size generally decreased with unfolding, where the increase represented an anomaly, which might be due to the selective unfolding of certain domains. The anomaly is marked red for BSA and green for HbAo.
cuvette of 1.5 mL (pathlength 1 cm). Before the dynamic light scattering (alternatively known as photon correlation spectroscopy) study, protein samples (typical concentration ~10Ag/mL) were passed through a 2-Am filter The operating procedure was programmed (using the DTS software supplied with the instrument) such that there are an average of 25 runs, each run being averaged for 15 seconds, with equilibration time 3 minutes at 258C. A particular hydrodynamic diameter (dh) is evaluated several times and the result is presented in terms of a distribution of the hydration diameter. Protein unfolding studies Protein was subjected to denaturing stress by adding 1 to 8 M urea as described earlier [7]. For each concentration fluorescence was measured with excitation at 275 and 290 nm to monitor the tyrosine and tryptophan fluorescence. A similar experiment was repeated with GNPs grown on such proteins. As a control, protein-free GNPs were also added to the unfolded protein. The experiment was repeated with both hemoglobin (Hb) and serum albumin [9]. Results and discussion
Dynamic light scattering
GNP-mediated protein unfolding via intermediate formation
The Nano-ZS (Malvern) instrument (5 mW HeNe laser E, 632 nm) was used for this purpose. The sample was poured into a DTS0112 low-volume disposable sizing
Figure 1, A and B shows the GNP colors developed at different stages of unfolding for BSA and Hb, respectively. The lower panel of Figure 1 shows the intensity ratio of the
Notes to Fig 1. Shift in plasmon resonance due to the formation of smaller particles with the increase in unfolding of the BSA. As we visualized they were photographed and were set in the increasing order of urea concentration. The lower panel shows the mean red pixel to mean blue pixel ratio of the digitized images of the NP-protein suspension. A, BSA. B, HbAo (normal human hemoglobin).
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
17
Fig 3. The dh is plotted against the plasmon resonance. Comparison of BSA and HbAo shows that the native state of the latter is confined to a red-shifted regime, and the HbAo has a wider span of plasmon resonance.
red plane and the blue plane. Interestingly, along with the unfolding there is a corresponding shift of particle size (Figure 2). Although there is an overall decrease in the particle size in the presence of higher urea concentration, there are certain intermediate unfolding stages that show anomalous increases in size. Interestingly for BSA, the concentrations at which such an increase in size is observed (1 M, 4-5 M, and 7 M) coincide with the concentrations that favor the formations of folding intermediates [10]. For Hb the change of Stokes diameter with unfolding is more or less gradual, except at 4 to 5 M urea. Interestingly, in our earlier work we reported a reversal of charge in Hb in this folding regime [7]. The image analysis and particle size variation described in Figures 1 and 2 resemble the unfolding patterns studied described in our earlier work [7]. The molecular ruler-like use of GNPs as described in Reinhard et al. [11] thus seems to be a distinct possibility, unfolding being associated with decrease of nanoparticle size and intermediate formation, exhibiting abrupt color changes. This is particularly obvious if we consider the Stokes diameter plotted against plasma resonance (Figure 3) that shows an overall decrease of dh with urea, with abrupt intermediate changes. The plasmon resonance versus
particle diameter profile is different for the two proteins. The difference in protein template thus influences the formation of the GNPs. Higher particle size with less red shift cannot be easily explained unless one assumes specific protein-dependent self-assembly of such particles. It seems that GNPs serve as reporters for the unfolding process. The onset of larger GNP formation at lower extent of unfolding may suggest that there is a correspondence with the molten-globule state that occurs only under low unfolding conditions. There is a well-known bypass of such states to the aggregative phase. The reporter nature of GNPs is further confirmed by the sudden alteration in particle size at denaturant concentrations that correspond to formation of folding intermediates. An insight into the folding process The most important observation that seemed to have a general validity in the heme as well as in the nonheme model proteins (Hb and BSA) is the strong scaling of the plasmon resonance (which in turn depends on the particle size) on the folding state of the protein. The tryptophan fluorescence of BSA and Hb (Figure 4, A and B) in turn had a similar dependence on the plasmon resonance (compare
18
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
Figures 3 and 4). It should be noted that in the case of BSA there was no appreciable change of fluorescence intensity when excitation was shifted from tyrosine maxima (275 nm) to tryptophan maxima (295 nm). In the case of Hb a differential alteration in the emission intensity at 335 nm was, however, observed. The emission profile was strongly correlated with unfolding as expected. The correspondence of the folding pathway as mapped by tryptophan or tyrosine, however, depends on the foldingdictated reorientation of the said residues. The difference with the GNP-sensed changes may be due to the higher sensitivity of the latter with respect to the foldingassociated changes. Protein-bound GNPs as quenchers of tyrosine fluorescence The increase in emission intensity (see Figure 4) [12], cannot be observed upon addition of preformed GNPs. Thus, quenching by seeded GNPs in the interaction with tyrosine can be ruled out. On the contrary, protein-free GNPs did not result in this level of attenuation of fluorescence. It is possible that complex formation of GNPs with tyrosine [4] severely affects this fluorescence. Mechanism of GNP size variation in the folding route The conditions that delimit the size of GNPs include the ionic strength and the reducing effect of the immediate environment (the template on which the GNPs are seeded). The question why the unfolding leads to formation of smaller sized GNPs may be answered in terms of exposure of hydrophobic groups that subject the GNP seeds to lower ionic strengths. Both the exposure of hydrophobic groups with unfolding and their concealment are well known in the literature. This conjecture may be proved by the observation that unfolding by any ionic denaturant masks this effect. Thus, the said effect can be realized only if the unfolding is mediated by a nonionic denaturant. Also possibly playing an important role is the modification of tyrosine residues. This may explain the quenching of tyrosine fluorescence described in the preceding section. Again, the exposure of the -SH groups (present in the cysteine residue) may facilitate the formation of a higher number of GNP seeds, which may ultimately limit their size. GNP-based identification of Hb variants The method for detection of different unfolding stages seemed applicable in classifying variants with aberrant folding behavior. Hb from thalassemia patients can be a rich source of such variants, some defective in chain composition and some containing mutations in certain positions in the h chain [8]. In Figure 5 the GNP-mediated color formation in unfolded variants is compared under denaturing conditions (4 M urea). Some variants have an intense bluish tint, implying larger particle formation. The reason we chose the denaturing condition was to avoid precipitation of certain variants due to unregulated formation of GNPs (with larger size).
Fig 4. Change in tryptophan and tyrosine fluorescence emission maxima. BSA (A) and (B) hemoglobin under different unfolded conditions induced by urea. The concentration of urea was taken along the x-axis, whereas the fluorescence intensity was taken along the y-axis. The emission was measured at 305 nm when excited at 275 nm for tyrosine (-o-) and at 335 nm when excited at 295 nm (-5-).
We have recently proposed a method to compare Hb variants depending on their interaction with copper nanoparticles [13]. The present approach may be considered as a simple classifier of folding intermediates and variants with defects in folding properties. Unlike the previous work, which was based on the clustering property of the nanoparticles that led to differential precipitation of the Hb variants, in this work a color-based sensing of the variants has been proposed. The simple observation that unfolding would lead to formation of smaller particles, except at conditions at which the folding intermediate is formed, may serve as an aid to characterize the folding route. Similarly the success of the color-based method to screen mutants or variants implies that defects in folding may serve as an important marker for a protein variant, and the difference can be sensed by nanoparticles.
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
19
Fig 5. Comparison of Hb variants. In each case unfolded (4M urea) proteins were compared. The E h variant was more bluish, as revealed from the diagram, and this was followed by the h major and h trait. The normal had the maximal red color.
Acknowledgments The authors thank Prof. U. Chaudhuri for providing the blood samples. They also thank Dr. Prabir Lahiri IHTM, Prof. Chanchal Mitra, Central University Hyderabad, and Mr. Ramaranjan Bhattacharya IACS for helpful suggestions. References [1] Chah S, Hammond MR, Zare RN. Old nanoparticles as a colorimetric sensor for protein conformational changes. Chem Biol 2005;12:323 - 8. [2] Gole A, et al. Pepsin-gold colloid conjugates: preparation, characterization, and enzymatic activity. Langmuir 2001;17:1674 - 9. [3] Mirkin CA, Letsinger RL, Mucic RC, Storhoff JJ. A DNA-based method for rationally assembling nanoparticles into macroscopic materials. Nature 1996;382:607 - 9. [4] Si S, Bhattacharjee RR, Banerjee A, Mandal TK. A mechanistic and kinetic study of the formation of metal nanoparticles by using synthetic tyrosine-based oligonucleotides. Chem Eur J 2006;12:1256 - 65. [5] Zhou Y, et al. Preparation of a novel core-shell nanostructured gold colloid-silk fibroin bioconjugate by the protein in situ redox technique at room temperature. Chem Commun 2001;2518 - 9. [6] Selvakannan PR, et al. Synthesis of aqueous Au core-Ag shell nanoparticles using tyrosine as a pH-dependent reducing agent and
[7]
[8]
[9]
[10]
[11]
[12]
[13]
assembling phase-transferred silver nanoparticles at the air-water interface. Langmuir 2004;20:7825 - 36. Bhattacharya J, et al. Unfolding of hemoglobin variants-insights from urea gradient gel electrophoresis photon correlation spectroscopy and zeta potential measurements. Anal Chim Acta 2004;522: 207 - 14. George PP, et al. Improvements in the HbVar database of human hemoglobin variants and thalassemia mutations for population and sequence variation studies. Nucleic Acids Res 2004;32:D537- 41. Anna K, James MN, David EG. The effect of azithromycin and clarithromycin on ex vivo interleukin-8 (IL-8) release from whole blood and IL-8 production by human alveolar macrophages. J Antimicrob Chemother 2001;47:867 - 70. Tayyab S, Sharma N, Khan M. Use of domain-specific ligands to study urea-induced unfolding of bovine serum albumin. Biochem Biophys Res Commun 2000;277:83 - 8. Reinhard BM, Siu M, Agarwal H, Alivisatos AP, Liphardt J. Calibration of dynamic molecular rulers based on plasmon coupling between gold nanoparticles. Nano Lett 2005;5:2246 - 52. Rosangela I, Wilker C, Leandro RSB, Mauricio SB. Effect of urea on bovine serum albumin in aqueous and reverse micelle environments investigated by small-angle X-ray scattering, fluorescence and circular dichroism. Braz J Phys 2004;34:58 - 63. Bhattacharya J, Choudhuri U, Siwach O, Sen P, Dasgupta AK. Interaction of hemoglobin and copper nanoparticle-implications in hemoglobinopathy. Nanomedicine 2006;2:191-8.
Clinical Nanomedicine
Gold nanoparticle-based tool to study protein conformational variants: implications in hemoglobinopathy Jaydeep Bhattacharya, MTech, Sinu Jasrapuria, MSc, Tapan Sarkar, MSc, Ranjita GhoshMoulick, MSc, Anjan Kr. Dasgupta, PhD4 Department of Biochemistry, Calcutta University, Calcutta, India Received 13 April 2006; accepted 24 October 2006
Abstract
The size of gold nanoparticles is shown here to gradually decrease if it is allowed to grow on a protein template, and the protein is subjected to unfolding by a nonionic denaturant. The correlation between size of the gold nanoparticle formed and the plasmon frequency observed remains linear, except at stages where protein folding intermediates are formed. Higher population of exposed tyrosine residues, number of sulfhydryl groups of the protein, and the overall exposition of the inner hydrophobic core may lead to the generation of smaller particles. The method provides a simple colorimetric sensing of protein conformation and has been tested for both nonheme and heme proteins (hemoglobin and bovine serum albumin). Similarly, protein variants with defects in folding (caused by subunit misassembly or mutation) can also be classified. Possible application of this approach in hemoglobinopathy (e.g., thalassemia carrier detection) is discussed in the text. D 2007 Published by Elsevier Inc.
Key words:
Gold nanoparticle; Plasmon resonance; Folding; Hemoglobinopathy; Hemoglobin; Serum albumin
Gold nanoparticles (GNPs) have been recently used in sensing folding states of proteins [1]. The observation has a wide range of implications in protein science, because it provides a simple colorimetric tool to follow the thermodynamic transitions occurring at different scales. The extent of reduction and the geometry of the reducing surface are known to play important roles in dictating the size of the emergent nanoparticles. The mechanism of the sensing technique relies on the size of the GNP, which in turn depends on the immediate reducing environment. A standard method of preparation of GNPs is sodium borohydride- or citrate-mediated reduction of aurochloric acid (HAuCl4). The particles thus produced should have a stabilized geometrical configuration to be
No conflict of interest was reported by the authors of this paper. 4 Corresponding author. Calcutta University, Calcutta, India. E-mail address: [email protected] (A.K. Dasgupta). 1549-9634/$ – see front matter D 2007 Published by Elsevier Inc. doi:10.1016/j.nano.2006.10.159
useful as sensors. If a protein template is present the reducing environment is altered. Amino groups or exposed thiol groups [2,3] present on the peptide surface can offer an environment that may facilitate emergence of size-stabilized particles. Again, exposed tyrosines may aid the reduction process, where an electron from the tyrosinate ion of the peptide is transferred to the metal ion at basic pH through the formation of a tyrosyl radical. The latter is converted to its dityrosine form during the reaction [4]. Recently, Zhou et al. prepared gold-silk fibroin core-shell nanoparticles by using tyrosine residues as both reducing and stabilizing agents, but they did not report the detailed nanoparticle formation mechanism [5]. Sastry and colleagues reported silver ion reduction with tyrosine at high pH [6]. In the investigation discussed here, a controlled alteration of the reducing surface was enforced by denaturants. Partial exposure of amino acid residues as a result of unfolding led to emergence of nanoparticles with a graded size distribution. The technique becomes particularly useful in sensing
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
15
16
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
the folding states of protein under identical buffering condition in response to denaturing stress. It may be important to cite Zare and co-workersT demonstration [1] once again. Because the unfolding was initiated by pH change the redox state of the reaction mixture might alter. It is tempting to ask whether the different colors are true representatives of different folding states or representatives of differential chemical activity of reducing agents originating in altered pH environments. We chose the nonionic denaturant urea to unfold the protein, so that the redox environment of the reaction mixture would not be disturbed. Eventually we could obtain a condition under which the unfolding studied using alternative methods and reported earlier [7] could be sensed by the plasmon frequency changes (as also reflected by size change of the GNPs). Methods Protein sample preparation Blood samples were collected in ethylenediamine tetraacetate (EDTA) vials from normal healthy persons. The whole blood was centrifuged at 3,000 g for 10 minutes and washed three times in normal saline. Hemolysis was carried out by mixing 100 AL of packed erythrocytes in 500 AL of lysis buffer (2.5 M NaCl, 100 mM EDTA-Na salt, 1% Triton X-100, and 10% dimethyl sulfoxide) [8]. The solution was applied to a Sephadex G-50 column equilibrated with 0.05 M phosphate buffer pH 7.0. Elution was carried out with 0.05 M phosphate buffer pH 7.0 at a flow rate of 25 mL/h. BBSA (10 mg/mL) was dissolved in 0.1 M phosphate buffer. Preparation of protein-free GNPs GNPs were prepared by the reduction of HAuCl4) by sodium borohydride. The normal reduction process was carried out according to the standard protocol. The size of GNPs obtained by this process was 14 nm [1]. Preparation of GNPs using proteins as templates The combined reducing property of sodium borohydride and the tyrosine of the protein HbA0 and BSA were used to reduce HAuCl4 to form GNPs. The major change that was made in this process was that the concentration of the sodium borohydride was kept low enough that it could not alone reduce HAuCl4 to form GNPs; yet the combined effect of the two reducing agents was capable of producing GNPs.
Fig 2. The mean hydrodynamic diameter (dh) of GNPs is plotted against urea concentration. The arrow indicates position where the folding intermediates may be formed. The size generally decreased with unfolding, where the increase represented an anomaly, which might be due to the selective unfolding of certain domains. The anomaly is marked red for BSA and green for HbAo.
cuvette of 1.5 mL (pathlength 1 cm). Before the dynamic light scattering (alternatively known as photon correlation spectroscopy) study, protein samples (typical concentration ~10Ag/mL) were passed through a 2-Am filter The operating procedure was programmed (using the DTS software supplied with the instrument) such that there are an average of 25 runs, each run being averaged for 15 seconds, with equilibration time 3 minutes at 258C. A particular hydrodynamic diameter (dh) is evaluated several times and the result is presented in terms of a distribution of the hydration diameter. Protein unfolding studies Protein was subjected to denaturing stress by adding 1 to 8 M urea as described earlier [7]. For each concentration fluorescence was measured with excitation at 275 and 290 nm to monitor the tyrosine and tryptophan fluorescence. A similar experiment was repeated with GNPs grown on such proteins. As a control, protein-free GNPs were also added to the unfolded protein. The experiment was repeated with both hemoglobin (Hb) and serum albumin [9]. Results and discussion
Dynamic light scattering
GNP-mediated protein unfolding via intermediate formation
The Nano-ZS (Malvern) instrument (5 mW HeNe laser E, 632 nm) was used for this purpose. The sample was poured into a DTS0112 low-volume disposable sizing
Figure 1, A and B shows the GNP colors developed at different stages of unfolding for BSA and Hb, respectively. The lower panel of Figure 1 shows the intensity ratio of the
Notes to Fig 1. Shift in plasmon resonance due to the formation of smaller particles with the increase in unfolding of the BSA. As we visualized they were photographed and were set in the increasing order of urea concentration. The lower panel shows the mean red pixel to mean blue pixel ratio of the digitized images of the NP-protein suspension. A, BSA. B, HbAo (normal human hemoglobin).
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
17
Fig 3. The dh is plotted against the plasmon resonance. Comparison of BSA and HbAo shows that the native state of the latter is confined to a red-shifted regime, and the HbAo has a wider span of plasmon resonance.
red plane and the blue plane. Interestingly, along with the unfolding there is a corresponding shift of particle size (Figure 2). Although there is an overall decrease in the particle size in the presence of higher urea concentration, there are certain intermediate unfolding stages that show anomalous increases in size. Interestingly for BSA, the concentrations at which such an increase in size is observed (1 M, 4-5 M, and 7 M) coincide with the concentrations that favor the formations of folding intermediates [10]. For Hb the change of Stokes diameter with unfolding is more or less gradual, except at 4 to 5 M urea. Interestingly, in our earlier work we reported a reversal of charge in Hb in this folding regime [7]. The image analysis and particle size variation described in Figures 1 and 2 resemble the unfolding patterns studied described in our earlier work [7]. The molecular ruler-like use of GNPs as described in Reinhard et al. [11] thus seems to be a distinct possibility, unfolding being associated with decrease of nanoparticle size and intermediate formation, exhibiting abrupt color changes. This is particularly obvious if we consider the Stokes diameter plotted against plasma resonance (Figure 3) that shows an overall decrease of dh with urea, with abrupt intermediate changes. The plasmon resonance versus
particle diameter profile is different for the two proteins. The difference in protein template thus influences the formation of the GNPs. Higher particle size with less red shift cannot be easily explained unless one assumes specific protein-dependent self-assembly of such particles. It seems that GNPs serve as reporters for the unfolding process. The onset of larger GNP formation at lower extent of unfolding may suggest that there is a correspondence with the molten-globule state that occurs only under low unfolding conditions. There is a well-known bypass of such states to the aggregative phase. The reporter nature of GNPs is further confirmed by the sudden alteration in particle size at denaturant concentrations that correspond to formation of folding intermediates. An insight into the folding process The most important observation that seemed to have a general validity in the heme as well as in the nonheme model proteins (Hb and BSA) is the strong scaling of the plasmon resonance (which in turn depends on the particle size) on the folding state of the protein. The tryptophan fluorescence of BSA and Hb (Figure 4, A and B) in turn had a similar dependence on the plasmon resonance (compare
18
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
Figures 3 and 4). It should be noted that in the case of BSA there was no appreciable change of fluorescence intensity when excitation was shifted from tyrosine maxima (275 nm) to tryptophan maxima (295 nm). In the case of Hb a differential alteration in the emission intensity at 335 nm was, however, observed. The emission profile was strongly correlated with unfolding as expected. The correspondence of the folding pathway as mapped by tryptophan or tyrosine, however, depends on the foldingdictated reorientation of the said residues. The difference with the GNP-sensed changes may be due to the higher sensitivity of the latter with respect to the foldingassociated changes. Protein-bound GNPs as quenchers of tyrosine fluorescence The increase in emission intensity (see Figure 4) [12], cannot be observed upon addition of preformed GNPs. Thus, quenching by seeded GNPs in the interaction with tyrosine can be ruled out. On the contrary, protein-free GNPs did not result in this level of attenuation of fluorescence. It is possible that complex formation of GNPs with tyrosine [4] severely affects this fluorescence. Mechanism of GNP size variation in the folding route The conditions that delimit the size of GNPs include the ionic strength and the reducing effect of the immediate environment (the template on which the GNPs are seeded). The question why the unfolding leads to formation of smaller sized GNPs may be answered in terms of exposure of hydrophobic groups that subject the GNP seeds to lower ionic strengths. Both the exposure of hydrophobic groups with unfolding and their concealment are well known in the literature. This conjecture may be proved by the observation that unfolding by any ionic denaturant masks this effect. Thus, the said effect can be realized only if the unfolding is mediated by a nonionic denaturant. Also possibly playing an important role is the modification of tyrosine residues. This may explain the quenching of tyrosine fluorescence described in the preceding section. Again, the exposure of the -SH groups (present in the cysteine residue) may facilitate the formation of a higher number of GNP seeds, which may ultimately limit their size. GNP-based identification of Hb variants The method for detection of different unfolding stages seemed applicable in classifying variants with aberrant folding behavior. Hb from thalassemia patients can be a rich source of such variants, some defective in chain composition and some containing mutations in certain positions in the h chain [8]. In Figure 5 the GNP-mediated color formation in unfolded variants is compared under denaturing conditions (4 M urea). Some variants have an intense bluish tint, implying larger particle formation. The reason we chose the denaturing condition was to avoid precipitation of certain variants due to unregulated formation of GNPs (with larger size).
Fig 4. Change in tryptophan and tyrosine fluorescence emission maxima. BSA (A) and (B) hemoglobin under different unfolded conditions induced by urea. The concentration of urea was taken along the x-axis, whereas the fluorescence intensity was taken along the y-axis. The emission was measured at 305 nm when excited at 275 nm for tyrosine (-o-) and at 335 nm when excited at 295 nm (-5-).
We have recently proposed a method to compare Hb variants depending on their interaction with copper nanoparticles [13]. The present approach may be considered as a simple classifier of folding intermediates and variants with defects in folding properties. Unlike the previous work, which was based on the clustering property of the nanoparticles that led to differential precipitation of the Hb variants, in this work a color-based sensing of the variants has been proposed. The simple observation that unfolding would lead to formation of smaller particles, except at conditions at which the folding intermediate is formed, may serve as an aid to characterize the folding route. Similarly the success of the color-based method to screen mutants or variants implies that defects in folding may serve as an important marker for a protein variant, and the difference can be sensed by nanoparticles.
J. Bhattacharya et al. / Nanomedicine: Nanotechnology, Biology, and Medicine 3 (2007) 14–19
19
Fig 5. Comparison of Hb variants. In each case unfolded (4M urea) proteins were compared. The E h variant was more bluish, as revealed from the diagram, and this was followed by the h major and h trait. The normal had the maximal red color.
Acknowledgments The authors thank Prof. U. Chaudhuri for providing the blood samples. They also thank Dr. Prabir Lahiri IHTM, Prof. Chanchal Mitra, Central University Hyderabad, and Mr. Ramaranjan Bhattacharya IACS for helpful suggestions. References [1] Chah S, Hammond MR, Zare RN. Old nanoparticles as a colorimetric sensor for protein conformational changes. Chem Biol 2005;12:323 - 8. [2] Gole A, et al. Pepsin-gold colloid conjugates: preparation, characterization, and enzymatic activity. Langmuir 2001;17:1674 - 9. [3] Mirkin CA, Letsinger RL, Mucic RC, Storhoff JJ. A DNA-based method for rationally assembling nanoparticles into macroscopic materials. Nature 1996;382:607 - 9. [4] Si S, Bhattacharjee RR, Banerjee A, Mandal TK. A mechanistic and kinetic study of the formation of metal nanoparticles by using synthetic tyrosine-based oligonucleotides. Chem Eur J 2006;12:1256 - 65. [5] Zhou Y, et al. Preparation of a novel core-shell nanostructured gold colloid-silk fibroin bioconjugate by the protein in situ redox technique at room temperature. Chem Commun 2001;2518 - 9. [6] Selvakannan PR, et al. Synthesis of aqueous Au core-Ag shell nanoparticles using tyrosine as a pH-dependent reducing agent and
[7]
[8]
[9]
[10]
[11]
[12]
[13]
assembling phase-transferred silver nanoparticles at the air-water interface. Langmuir 2004;20:7825 - 36. Bhattacharya J, et al. Unfolding of hemoglobin variants-insights from urea gradient gel electrophoresis photon correlation spectroscopy and zeta potential measurements. Anal Chim Acta 2004;522: 207 - 14. George PP, et al. Improvements in the HbVar database of human hemoglobin variants and thalassemia mutations for population and sequence variation studies. Nucleic Acids Res 2004;32:D537- 41. Anna K, James MN, David EG. The effect of azithromycin and clarithromycin on ex vivo interleukin-8 (IL-8) release from whole blood and IL-8 production by human alveolar macrophages. J Antimicrob Chemother 2001;47:867 - 70. Tayyab S, Sharma N, Khan M. Use of domain-specific ligands to study urea-induced unfolding of bovine serum albumin. Biochem Biophys Res Commun 2000;277:83 - 8. Reinhard BM, Siu M, Agarwal H, Alivisatos AP, Liphardt J. Calibration of dynamic molecular rulers based on plasmon coupling between gold nanoparticles. Nano Lett 2005;5:2246 - 52. Rosangela I, Wilker C, Leandro RSB, Mauricio SB. Effect of urea on bovine serum albumin in aqueous and reverse micelle environments investigated by small-angle X-ray scattering, fluorescence and circular dichroism. Braz J Phys 2004;34:58 - 63. Bhattacharya J, Choudhuri U, Siwach O, Sen P, Dasgupta AK. Interaction of hemoglobin and copper nanoparticle-implications in hemoglobinopathy. Nanomedicine 2006;2:191-8.