Mercuration of apo-α-lactalbumin: binding of Hg²⁺ followed by protein-mediated nanoparticle formation

Abstract

Nanoparticles and nanocrystals of mercury are formed when Hg²⁺ salt reacts with apo-α-lactalbumin (apo-α-LA). Reduction followed by nanoparticle formation is further augmented by the protein, as it also acts as a coating agent. The initial interaction of Hg²⁺ with apo-α-LA was demonstrated by changes observed in absorption, emission, and CD spectroscopy, where the latter technique also detected structural changes in the protein. Such structural changes are expected when Hg²⁺ binds to the protein, and therefore, the binding was determined by isothermal titration calorimetry (ITC). The binding was further proven by MALDI, which showed mercurated species of protein with a Gaussian distribution exhibiting a weighted average of 6 and 9 Hg²⁺ ions bound to protein when apo-α-LA was treated with 10 and 100 equivalents, respectively. The molecular dynamics studies revealed the binding of Hg²⁺ ions followed by the structural changes that occurred in the protein. The reaction between Hg²⁺ and apo-α-LA yields non-crystalline nanoparticles at lower molar ratios of Hg²⁺ and crystalline ones at higher molar ratios. The existence of both of these nanoparticles was proven by extensive TEM studies, and the mercury nanocrystals were further studied using fluorescence microscopy. X-ray photoelectron spectroscopy demonstrated that the protein has the ability to convert Hg²⁺ to Hg⁰, and the resultant Hg⁰ cluster is known to be less harmful than Hg²⁺ to the organism. All of these studies support the use of apo-α-LA in the form of nanoparticles and nanocrystals to detoxify Hg²⁺.

---

Introduction

Since ancient days, metal nanoparticles have been of interest because of their technological applications.¹ Several transition metals including silver and gold have been studied to discern their nanocharacteristics.^2,3 Although mercury has special properties and characteristics among the metallic elements, its nanostudies are limited.¹ The toxic properties of mercury are responsible for diseases such as acrodynia, Hunter-Russell syndrome, Minamata, and neurodegenerative disorders.⁴ Even a 5 μM Hg²⁺ solution can inhibit the elongation of Oryza sativa seedlings.⁵ The Hg²⁺ ion is hazardous because of its affinity to interact with protein residues such as cysteine, histidine, aspartic acid, and glutamic acid.^6,7 The interaction of Hg²⁺ with free thiols of protein helps to decrease the toxicity of the metal ion by converting Hg²⁺ to Hg⁰. The development of techniques for easy and environmentally benign synthesis of stable mercury nanocrystals is currently a great challenge.¹ While dental amalgams that contain mercury are well known in the literature, the generation of nanostructured mercury requires a harsh method.^1,8

In order to understand the effect of mercuration on proteins, apo-α-lactalbumin (apo-α-LA) was chosen because of its small size and its exposed residues of Glu, Asp, His, and Tyr, and also because of the prior knowledge gathered through studies carried out with Pb²⁺, Mg²⁺, Zn²⁺, Cu²⁺ and Al³⁺.^9–11 It has been previously demonstrated that the side chains of Glu, Asp, His, and Tyr residues present in Aβ peptides and in proteins such as α-lactalbumin, human serum albumin, and β-lactoglobulin interact with various metal ions, resulting in aggregation and fiber formation.^11–14 During the interaction of metal ions and protein, the protein often reduces the bound ions to form metallic species and/or nanocrystals.² The interaction between α-LA and Hg²⁺ results in the formation of Hg⁰, which is less toxic and is well known for its ability to detoxify Hg²⁺ in some bacteria.⁶ Therefore, the aim of the present study is not only to demonstrate the binding of Hg²⁺ to apo-α-LA, but also to study the speciation and the ability of the protein to reduce the Hg²⁺ toxicity by converting this ion into a less toxic form, viz., Hg⁰, resulting in nanoparticles and nanocrystals. Different techniques were adopted, such as thermodynamics, spectroscopy, and microscopy coupled with molecular dynamics modeling to obtain the necessary data. Therefore, for the first time, we have demonstrated that mercury nanocrystals form under mild conditions wherein a protein plays an important role.

Results and discussion

Interaction between apo-α-LA and Hg²⁺

The interaction of Hg²⁺ and apo-α-LA was demonstrated based on absorption, emission, and CD spectroscopy. Apo-α-LA alone shows two characteristic absorption bands at 210 and 280 nm with a shoulder at 245 nm in 10 mM Tris buffer at pH = 7.4. At a low concentration (1–10 equivalents) of Hg²⁺, no significant changes were observed. However, upon addition of 100 equivalents, the absorbance of the 280 nm band increases from 0.3 to 1.4 and shows a prominent band at 300 nm along with a shoulder centered at approximately 350 nm (Fig. 1a), and such changes have been reported to be characteristic of metal ion binding.^15,16 The mercury nanocrystal exhibits a characteristic surface plasmon resonance (SPR) band at 290 nm (Fig. 1b). The nanoparticle formation was also confirmed by the TEM studies. However, the absorbance of the 210 nm band decreases exponentially as a function of added [Hg²⁺] and is associated with a 25–30 nm redshift in the peak position that is attributable to the interaction of Hg²⁺ with the peptide bonds of α-LA. Indeed, such interactions have been reported in the literature.¹⁷

Fig. 1 (a) Absorption titration of 10 μM of {apo-α-LA} in 10 mM Tris buffer at pH = 7.4 upon addition of 1 to 100 equivalents of Hg²⁺. Inset: The absorbance vs. molar ratio is plotted at 210 () and 280 () nm. (b) Subtraction of the α-LA spectrum from the {apo-α-LA + xHg²⁺} spectra where x = 10 (), 20 (), 50 (), 80 () and 100 () equivalents. (c) Fluorescence titration of 3 μM of {apo-α-LA} in 10 mM Tris buffer at pH = 7.4 upon addition of 1 to 100 equivalents of Hg²⁺. Inset: I/I₀ vs. molar ratio plot for the 330 nm band. (d and e) Far-UV and near-UV CD spectra obtained during the titration of {apo-α-LA} () with 1 (), 3 (), 5 (), 10 () and 100 () equivalents of Hg²⁺, respectively.

The fluorescence in apo-α-LA arises from all the aromatic amino acids, viz., Trp, Tyr, and Phe. Most of these aromatic amino acids are exposed, and therefore, the fluorescence emission is primarily from Trp 104, which is buried in the protein. The fluorescence intensity of apo-α-LA observed at 335 nm decreases upon addition of Hg²⁺. An increase in the [Hg²⁺] up to 100 equivalents caused a decrease in the fluorescence intensity that was accompanied by a redshift of approximately 5 nm, resulting in the translocation of the Trp residue to the hydrophilic region due to the changes in the microenvironment (Fig. 1c).

Far UV CD is a well known tool in establishing protein conformation and nanoparticle formation.¹⁸ The far-UV CD spectra of apo-α-LA in the absence of Hg²⁺ shows two negative bands centered at 222 and 208 nm, which are characteristic of an α-helix. While the addition of [Hg²⁺] up to 10 equivalents brings only marginal changes, the addition of up to 100 equivalents brings a significant loss in the negative ellipticity of the band at 222 nm (Fig. 1d and e). This suggests that the addition of approximately 100 equivalents [Hg²⁺] destabilizes the secondary structure of apo-α-LA. Such structural changes favor the formation of nanoclusters of mercury-bound apo-α-LA, which is further supported by additional data in this study. Significant changes were noticed in the microenvironment of the aromatic amino acid residues, suggesting alterations in the tertiary structure of the protein in the presence of Hg²⁺ (Fig. 1e).

Binding of Hg²⁺ to apo-α-LA

The binding of mercury by apo-α-LA was demonstrated by ITC and MALDI mass spectrometry, and the species were modeled by MD computations. The binding of Hg²⁺ to α-LA was confirmed from ITC studies. Two moderate binding sites were observed by ITC with K₁ = (3.5 ± 1.5) × 10⁴ and K₂ = (4.3 ± 2.6) × 10⁴ M⁻¹ when the titration data was fitted using a sequential binding model (Fig. 2). Both sites were exothermic and enthalpically favorable. The binding of Hg²⁺ was modeled based on MD computations.

Fig. 2 ITC profile and best fit for the interaction of Hg²⁺ (3.5 mM) with {apo-α-LA} 250 μl, 70 μM in 10 mM Tris buffer at pH = 7.4. Upper panel: Raw data obtained from several consecutive injections of 2 μl of Hg²⁺ per injection. Lower panel: the points are from the experiment and were subtracted from the contribution of the appropriate control, and the solid line is the best fit.

In order to understand the binding features associated with Hg²⁺ binding to α-LA, MD simulations were carried out for 25 ns at the conditions mentioned in the experimental section. At lower equivalents of Hg²⁺, the MD study resulted in α-LA bound by three mercury ions through carboxylate side chain moieties (SI 01†) with no significant changes in the conformation of the protein. This is also evident from the CD study carried out at this concentration. At higher equivalents, this resulted in the binding of six Hg²⁺ ions, viz., Hg1 to Hg6, to the surface of α-LA, as can be seen from Fig. 3.

Fig. 3 Results of the MD studies carried out between α-LA and 50 equivalents of Hg²⁺: (a) snapshots taken after a 25 ns MD simulation shows the interaction of six mercury ions with α-LA. The binding cores for individual mercury centers are shown in: (b) for Hg1 and Hg2; (c) for Hg3; (d) for Hg4; (e) for Hg5; and (f) for Hg6. (g) Overlay structure of the snapshot taken from a 25 ns MD simulation (in blue colour) and the crystal structure of α-LA (in yellow colour). Major structural changes that occurred in the α-LA are encircled in the corresponding amino acid residues (i) 100 to 111; (ii) 43 to 48; (iii) 57–60; (iv) 62–69; (v) 76 to 86.

At this concentration of Hg²⁺, the original Ca²⁺ site is disrupted due to the changes made to the structure of the protein, which were dependent on the number of equivalents of Hg²⁺ added; this was supported by CD studies. Significant structural changes were observed in the regions marked in Fig. 3g, wherein a striking variation is observed in the peptide region spanning from 100 to 111 amino acids, where the helical region is converted to a random coil. Thus, the MD results concur with those obtained from CD data, suggesting that although the protein conformation is unaltered at a low equivalent of Hg²⁺, the conformation is altered at a higher equivalent addition of Hg²⁺ due to the structural changes occurring in the protein.

The Hg²⁺-bound species of α-LA were determined by MALDI-TOF MS (Fig. 4). These experiments were carried out for a solution of α-LA treated with up to 100 equivalents of Hg²⁺. As the number of added Hg²⁺ equivalents increases to 10, the peak corresponding to the unbound protein decreases and the protein bound to one, two, and three ions of Hg²⁺ increases, as can be seen from Fig. 4b–d. The initial result is similar to the ITC and the MD results, where two and three bindings were observed, respectively. However, when the solution of the 10 equivalents was incubated for one day, the peak corresponding to the unbound protein completely vanished and resulted in a complex spectrum where peaks corresponding to 1 to 9 ions of Hg²⁺-bound protein (Fig. 4e) are seen. The intensity profile of this complex spectrum follows a Gaussian distribution, and the maximum peak intensity was observed for 6 and 7 Hg²⁺-bound species (Fig. 4e). A weighted average shows the presence of approximately 6 Hg²⁺ per protein molecule.

Fig. 4 MALDI-TOF spectra of {apo-α-LA + x equivalents of Hg²⁺}: (a) x = 0; (b) x = 2; (c) x = 5; (d) x = 10; (e) x = 10, but after one day of incubation; (f) x = 100. The number shown on each peak corresponds to the number of Hg²⁺ ions bound.

In the case of 100 equivalents of Hg²⁺, the MALDI experiment yielded a complex spectrum that is similar, with a maximum intensity found for 10 Hg²⁺-bound proteins and the weighted average shifted to 9 Hg²⁺ per protein. A similar pattern was observed even after one day of incubation of the 100 equivalent solution. Because all these mercurated α-LA species were observable by MALDI, they must have moderate binding strengths, as also supported by the ITC. The current study originally suggested that 2–3 binding sites are available for Hg²⁺; however, upon addition of a higher equivalent of Hg²⁺, changes occur in the protein conformation and structure that further favor the binding of more Hg²⁺ ions to the protein. Secondary and tertiary structural changes caused by Hg²⁺ binding were already shown by the CD studies. All these studies support the formation of a statistical mixture of an Hg²⁺-bound α-LA species.

Formation of mercury nanoclusters covered with apo-α-LA

As Hg²⁺-induced structural changes of the protein were observed, the nanoparticles and nanocrystal formation was addressed by transmission electron microscopy (TEM) (Fig. 5, SI 02–05†). Apo-α-LA (10 μM), upon interaction with 10 equivalents of Hg²⁺, forms a non-crystalline type of nanoparticle that is 10–20 nm size (Fig. 5a–d). At low equivalents of Hg²⁺, apo-α-LA does not exhibit any significant change in its secondary structure based on CD studies, and hence no exposure of tyrosine and cysteine is expected to occur. Even upon increasing the equivalents to 100, the size of the nanoparticles that formed remained almost the same. However, several new characteristics of the nanocrystals were noticed due to secondary structural changes that are expected in the protein at this concentration of Hg²⁺, resulting in the exposure of the Tyr and Cys residues and thereby supporting the formation of crystalline nanoparticles of mercury (Fig. 5e–k, SI 02†). Similar particles were observed with a sample of apo-α-LA with 100 equivalents of Hg²⁺ that was dialyzed (SI 03†).

Fig. 5 TEM micrographs (a and b) SAED; (c) EDX; (d) of {apo-α-LA + 10 equivalents of Hg²⁺}. TEM micrographs (e) high resolution TEM image of one Hg nanocrystal; (f) SAED; (g) EDX; (h) of {apo-α-LA + 100 equiv. Hg²⁺}; (i–k) High resolution TEM images of three different Hg nanocrystals; and (l and m) Size distribution histograms of a and e, respectively.

Comparatively, large particles with a polycrystalline and/or amorphous nature were found with mercury perchlorate(II) salts used as a control for comparison (SI 04†). ANS study also supported the exposure of Tyr and Cys in the protein upon Hg²⁺ addition (SI 05†).

Exposure of Tyr and Cys followed by the reduction of Hg²⁺ to Hg⁰ by apo-α-LA

The treatment of apo-α-LA by Hg²⁺ resulted in mercury-bound species, as determined by MALDI-TOF, which indicated the formation of nanoparticles that were confirmed by TEM. The reduction of Hg²⁺ by the protein was studied by X-ray photoelectron spectroscopy (XPS) for the apo-α-LA that was treated with 100 equivalents of Hg²⁺ (Fig. 6a, SI 06†). The spectra showed peaks at 98.9 and 101.9 eV, respectively, for Hg⁰ and Hg²⁺, and the assignment of those peaks is in agreement with that given in the literature.¹⁹ This shows that the precursor Hg²⁺ is reduced to a less toxic Hg⁰ that is possible by the oxidation of the phenol moieties of four tyrosine residues in apo-α-LA. This is further supported by absorption studies (Fig. 6b) wherein the bands observed at 280 and 210 nm for α-LA were shifted to 285 and 235 nm, respectively, with a shoulder at approximately 245 nm; all this clearly suggests the exposure of tyrosine moieties to the environment.²⁰

Fig. 6 (a) XPS spectra of Hg: α-LA () and its mercury-bound species (). (b) Absorption spectra of α-LA () and {α-LA + 100 equivalents of Hg²⁺} ().

Fluorescence microscopy studies of the Hg²⁺-bound apo-α-LA

The laser confocal fluorescence microscopy image of {apo-α-LA + 100 equivalents of Hg²⁺} shows green fluorescence that is not found with the protein alone, thereby suggesting the formation of nanocrystals or nanoparticles of small size (5–10 nm) (Fig. 7a and b). However, the {apo-α-LA + 100 equivalents of Hg²⁺} exhibits a broad emission band centered at 400 nm (Fig. 7c).

Fig. 7 Laser confocal fluorescence microscopy images in 10 mM Tris buffer, pH = 7.4: (a) apo-α-LA; (b) {apo-α-LA + 100 equivalents Hg²⁺}; (c) fluorescence spectra for {apo-α-LA} () and {apo-α-LA + 100 equivalents Hg²⁺} ().

Conclusions and correlations

This study shows for the first time the utilization of protein, apo-α-LA, in the formation of mercury nanoparticles and nanocrystals. Although it is known that heavy metals such as Hg²⁺ and Pb²⁺ cause protein aggregation, the molecular details of detoxification of Hg²⁺ by proteins and in turn its effect on the protein structure and function, particularly the aggregational behavior, are not well understood.⁷ The aggregation is known to be the result of the exposure of hydrophobic residues. Controlling hydrophobic interactions is a means to control the size of nanoparticles. This, in turn, is regulated by changing the Hg²⁺ and apo-α-LA concentrations. Different concentrations of Hg²⁺ were used to study the effect on apo-α-LA conformation that results in nanocrystal formation. To our knowledge, this is the first example of green synthesis of mercury nanocrystals that is affected by the reducing action of the protein, apo-α-LA. Such nanocrystal formation is due to the specific and non-specific interaction of Hg²⁺ with α-LA followed by reduction. While the two specific binding sites were proven by ITC titrations, the specific as well as non-specific bindings were shown by MALDI mass spectrometry, where each peak of the complex spectrum observed corresponds to the protein bound to one or more Hg²⁺ ions, with a maximum number of 15 Hg²⁺ ions being bound. A weighted average of this shows binding by at least 10 Hg²⁺ ions. The binding of Hg²⁺ with carboxylate, histidine, and tyrosine as well as cysteine (disulfide form) residues results in the exposure of hydrophobic patches followed by the loss of secondary structure that induces the formation of nanoclusters. These findings were supported by MALDI and TEM studies.

The nanocharacteristics of the protein-coated mercury depend on the concentration of Hg²⁺ ions. While non-crystalline nanoparticles were observed at lower molar ratios of Hg²⁺ ions, crystalline nanoparticles were observed at higher molar ratios. These nanocrystals were found to have the ability to convert highly toxic Hg²⁺ into the less toxic Hg⁰ form using the redox ability of different residues, including the tyrosine present in the protein. A recent report in the literature describes the synthesis of mercury nanocrystals carried out under harsh conditions consisting of nitric acid plus a polyvinyl alcohol mixture of mercury nitrate heated at 60–110 °C.¹ In contrast, the present method utilizes a simple and environmentally friendly synthesis of the mercury nanocrystals, wherein the protein acts as a reducing as well as capping agent, and thus detoxifies mercury.

Experimental section

Materials

Bovine apo-α-LA and mercury perchlorate were procured from the Sigma-Aldrich Chemical Co., USA. All the other chemicals used were from E. Merck, Germany. An apoform of lactalbumin (apo-α-LA) was used in all experiments. The concentration of apo-α-LA was calculated from the absorbance at 280 nm using the molar extinction coefficient of 28 500 M⁻¹ cm⁻¹.²¹

Nanoparticle preparation

To a requisite volume of 10 μM of apo-α-LA solution in 1 ml Eppendorf tubes, 1 and 10 μl of 100 mM mercury perchlorate (Hg²⁺) was added to give 1 : 10 and 1 : 100 molar ratios of protein to metal ion, which were maintained throughout all the studies reported here. The 1 ml solutions thus prepared were incubated in a shaking stirrer for 2 h, and these were used for further studies.

Absorption spectra

UV-visible absorption studies were performed using a Varian Cary 100 UV-spectrophotometer. A 10 μM cuvette concentration of protein was used for all experiments. The Hg²⁺ titrations of apo-α-LA were performed by continuous addition from 1 to 10 μl of the 10 mM and 100 mM Hg²⁺ stock solutions in such a way that the final volume of the solution was always maintained at 1 ml. A control was performed by titrating the Hg²⁺ with buffer and subtracting this from the {apo-α-LA + Hg²⁺} titrations. Only the subtracted data are shown as the final spectra in this study.

Emission spectra

Fluorescence studies were performed on a PerkinElmer LS 55 fluorescence spectrometer. The samples were prepared in the same manner as that given for the absorption studies. These samples were diluted three-fold so that the final protein concentration was 3 μM. The metal ion concentration was adjusted so that the molar ratio for all the studies remained the same. The resultant solutions were used for the fluorescence studies by exciting at 280 nm and measuring the spectra from 290 to 500 nm using a 5 nm slit both for the excitation and emission gates, with a scan speed of 150 nm min⁻¹. The control was performed with simple Hg²⁺ ion solution and was subtracted from the {apo-α-LA + Hg²⁺} titrations.

Matrix assisted laser desorption ionization (MALDI) mass spectrometry

The mass spectra of native and metallated proteins were recorded on an Autoflex III TOF/TOF MS (MALDI-TOF mass spectrometer, Bruker Daltonics Co.). α-Cyano-hydroxy-trans-cinnamic acid was used as the matrix for the MALDI-TOF. The MALDI-TOF instrument was calibrated with horse heart myoglobin (m/z values, 16 952) prior to the measurements.

XPS analysis

X-ray photoelectron spectroscopy (XPS) was performed in the analysis chamber of an Omicron XPS/UPS system. The base pressure of the chamber was 1.5 × 10⁻⁷ Pa. Aluminium K-alpha X-ray (1486.6 eV) was used as the excitation source. All the data were baseline corrected and plotted using Origin 7.5 software.

Isothermal titration calorimetry (ITC)

Calorimetric titrations were performed at 25 °C with a microcal isothermal titration calorimeter from MicroCal ITC₂₀₀ (Northampton, MA, USA) using 5 mM Hg²⁺. Successive additions of 2 μl of Hg²⁺ were separated by a 150 s interval to allow the peak resulting from the {apo-α-LA + Hg²⁺} interaction to return to the baseline. The ITC data were fitted with the Origin software package provided by MicroCal using curve fitting models of sequential sites.

Sample preparation for FEG-TEM

In order to carry out the microscopy studies, 10 μM of apo-α-LA was incubated with 10 and 100 equivalents of Hg²⁺. Each sample was then sonicated for 5 min. The samples were spread over carbon-coated copper grids with a 200 mesh, dried, and then analyzed with a JEOL JEM-2100F (FEG-TEM) electron microscope operating at 200 kV.

Binding by 8-anilinonaphthalene-1-sulfonate (ANS)

Fluorescence emission spectra for ANS were recorded from 400 to 600 nm by exciting the solutions at 365 nm. The excitation and emission slit widths were kept at 5 nm. 2–10 μl of a 2 mM ANS solution was added serially to a total volume of 1 ml, and the spectra were recorded. A blank titration was also performed with simple buffer as a control and was subtracted from the corresponding titration data.

Circular dichroism (CD) spectroscopy

Far-UV CD spectra were collected using a Jasco J-815 CD spectrometer. The path length of the cuvette was 2 mm. Apo-α-LA samples with 1–10 equivalent of Hg²⁺ were studied at 10 μM concentration throughout the CD spectra. Three scans were recorded with a scan speed of 100 nm min⁻¹ and were averaged.

Molecular dynamics (MD) simulations

The initial coordinates of α-LA were downloaded from the X-ray crystal structure of α-LA (PDB code: 1HFZ).²² The two independent simulation studies were carried out with 10 and 50 Hg²⁺ ions. The initial models for both simulations were made from the protein crystal structure with the following modifications: (a) the unbound water molecules were removed, (b) Ca²⁺ was removed by retaining the two coordinated water molecules, (c) hydrogen atoms were added using the XLEaP program, (d) Hg²⁺ ions were manually placed around the protein randomly using the Accelrys DS Visualizer before performing the simulations study. All MD simulations were performed under periodic boundary conditions (PBC) using an explicit solvent (TIP3P water molecules) environment with the Amber 10 software package employing ff03 force field parameters.^23,24 α-LA was solvated with 6760 and 7533 TIP3P²⁵ water molecules for the 10 and 50 Hg²⁺ simulation models, respectively. A truncated octahedral box was used for the calculation with an initial volume of 264 257 and 298 000 ų for the 10 and 50 Hg²⁺ simulation models, respectively. The charges for glutamate and aspartate residues were assigned as −1, lysine and arginine residues were set as +1, all N^ε2-positions of histidines were protonated, and Cl⁻ was added to neutralize the charge balance. The van der Waals radius (R, Å) and potential energy well depth (ε, kcal mol⁻¹) parameters for Hg(II) were 1.60 and 1.0, respectively. The isothermal–isobaric (NPT) ensemble was employed with PBC. The trajectories were saved for every 1 ps interval for further analysis, and processing of the trajectories was carried out using the PTRAJ software package. The Particle Mesh Ewald (PME) summation was used to treat long-range electrostatic interactions, and a 12 Å cutoff was used for the van der Waals interactions.²⁶ The SHAKE algorithm was used to constrain the bond lengths involving hydrogen atoms with a relative geometric tolerance of 0.0001. Constant pressure (1 atm) and temperature were maintained using the Berendsen weak-coupling algorithm²⁷ with a relaxation time constant of 2 ps and Langevin thermostat with a time constant for heat bath coupling of 1 ps, respectively.^28,29 All simulations were subjected to temperature, pressure, potential energy, and kinetic energy calculations as a function of time, and no significant changes were observed in these parameters.

Confocal laser scanning microscopy (CLSM)

The samples for the CLSM were prepared on the coverslip and air dried, and were then scanned under the laser source having an excitation of 250–360 nm using an Olympus IX. The images were analyzed using Fluoview Viewer software.

Acknowledgements

We dedicate this paper to Professor C.N.R. Rao on his 80th birthday. CPR acknowledges the financial support from DST, CSIR and DAE-BRNS. VKH acknowledges the IIT Bombay for RA ship. We thank IIT Bombay for TEM and MALDI-TOF. We thank BRAF (Bioinformatics Resources and Applications Facility) at Centre for Development of Advanced Computing (C-DAC), Pune, India for computational time.

References

1. G. V. Ramesh, M. D. Prasad and T. P. Radhakrishnan, Chem. Mater., 2011, 23, 5231–5236.
2. J. Xie, Y. Zheng and J. Y. Ying, J. Am. Chem. Soc., 2009, 131, 888–889.
3. A. K. Singh, R. Kanchanapally, Z. Fan, D. Senapati and P. C. Ray, Chem. Commun., 2012, 48, 9047–9049.
4. P. S. Harikumar, A. Dhruvan, V. Sabna and A. Babitha, J. Toxicol. Environ. Health, 2011, 3, 008–013.
5. M. Patra and A. Sharma, Bot. Rev., 2000, 66, 379–422.
6. H. Qian, L. Sahlman, P. O. Eriksson, C. Hambraeus, U. Edlund and I. Sethson, Biochemistry, 1998, 37, 9316–9322.
7. D. B. Veprintsev, E. A. Permyako, L. A. Kalinichenko and L. J. Berliner, Biochem. Mol. Biol. Int., 1996, 39, 1255–1265.
8. K. V. Katok, R. L. D. Whitby, T. Fukuda, T. Maekawa, I. Bezverkhyy, S. V. Mikhalovsky and A. B. Cundy, Angew. Chem., Int. Ed., 2012, 51, 2632–2635.
9. A. Saha and V. V. Yakovlev, J. Biophotonics, 2010, 3, 670–677.
10. P. A. Eugene and B. J. Lawrence, FEBS Lett., 2000, 473, 269–274.
11. J. F. Graveland-Bikker, I. A. T. Schaap, C. F. Schmidt and C. G. de Kruif, Nano Lett., 2006, 6, 616–621.
12. P. Faller, C. Hureau and O. Berthoumieu, Inorg. Chem., 2013, 52, 12193–12206.
13. T. Dudev and C. Lim, Chem. Rev., 2013, 1, 538–560.
14. B. Zappone, M. P. De Santo, C. Labate, B. Rizzutia and R. Guzzi, Soft Matter, 2013, 9, 2412–3241.
15. B. Roschitzki and M. Vasak, J. Biol. Inorg. Chem., 2002, 7, 6–616.
16. G. Meloni, P. Faller and M. Vasak, J. Biol. Chem., 2007, 282, 16068–16078.
17. S. M. Kelly and N. C. Price, Curr. Protein Pept. Sci., 2000, 1, 349–384.
18. Y. Yu, Z. Luo, C. S. Teo, Y. N. Tan and J. Xie, Chem. Commun., 2013, 49, 9740.
19. N. D. Hutson, B. C. Attwood and K. G. Scheckel, Environ. Sci. Technol., 2007, 41, 1747–1752.
20. E. P. Melo, M. R. Aires-Barros, S. M. B. Costa and J. M. S. Cabral, J. Biochem. Biophys. Methods, 1997, 34, 45–59.
21. A. Vanhooren, K. Vanhee, K. Noyelle, Z. Majer, M. Joniau and I. Hanssens, Biophys. J., 2002, 82, 407–417.
22. A. C. Pike, K. Brew and K. R. Acharya, Structure, 1996, 4, 691–703.
23. A. Case, T. A. Darden, T. E. Cheatham III, C. L. Simmerling, J. Wang, R. E. Duke, R. Luo, M. Crowley, R. C. Walker, W. Zhang, K. M. Merz, B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry, K. F. Wong, F. Paesani, J. Vanicek, X. Wu, S. R. Brozell, T. Steinbrecher, H. Gohlke, L. Yang, C. Tan, J. Mongan, V. Hornak, G. Cui, D. H. Mathews, M. G. Seetin, C. Sagui, V. Babin and P. A. Kollman, AMBER 10, University of California, San Francisco, CA, 2008.
24. Y. Duan, C. Wu, S. Chowdhury, M. C. Lee, G. Xiong, W. Zhang, R. Yang, P. Cipelak, R. Luo and T. Lee, J. Comput. Chem., 2003, 24, 1999–2012.
25. W. L. Jorgensen, J. Chandrasekhar, J. D. Madura, R. W. Impey and M. L. Klein, J. Chem. Phys., 1983, 79, 926–935.
26. T. Darden, D. York and L. Pedersen, J. Chem. Phys., 1993, 98, 10089–10092.
27. H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, A. DiNola and J. R. Haak, J. Chem. Phys., 1984, 81, 3684–3690.
28. B. P. Uberuaga, M. Anghel and A. F. Voter, J. Chem. Phys., 2004, 120, 6363–6374.
29. J. P. Ryckaert, G. Ciccotti and H. J. C. Berendsen, J. Comput. Phys., 1977, 23, 327–341.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra07156e

---