Quenching refers to any process which decreases the fluorescence intensity of a given substance. A variety of processes can result in quenching, such as excited state reactions, energy transfer, complex-formation and collisional quenching. Figure 1 shows the fluorescence spectra of HSA in the absence and the presence of nC₆₀ in phosphate buffer with an excitation wavelength of 280 nm. HSA displayed the maximum emission at a wavelength of 343 nm. The fluorescence intensity of HSA decreased with increasing concentration of nC₆₀ and the maximum emission wavelength was slightly red shifted. It is known that the shift of maximum emission wavelength corresponded to a polarity change around the chromophore residues. Red shift indicates that the microenvironment of fluorophors in HSA is changed after addition of nC₆₀.

For dynamic quenching, the quenching is an additional process that deactivates the excited state besides radiative emission through the collision between the quencher and fluorophore. As for HSA, the dependence of the emission intensity on quencher concentration [Q] is given by the Stern-Volmer equation: (1) F 0 / F = 1 + K q τ 0 [ Q ] = 1 + K S V [ Q ]where F and F₀ are the fluorescence intensity in the absence and in the presence of quencher, respectively; τ₀ is the lifetime of HSA; K_q is the bimolecular rate constant for the dynamic reaction of the quencher with the fluorophore; K_sv is called the Stern-Volmer constant.

For static quenching, the binding of a quencher (Q) to a protein (M) can be described with the following reaction: nQ + M ↔ MQ n

Then the binding constant (K) is given by (2) K = [ MQ n ] [ M ] [ Q ] n

The relation between the fluorescence intensity and concentration of quencher [ Q ] is given by (3) [ M ] 0 = [ M ] + [ MQ n ] (4) F 0 − F F = [ M ] 0 − [ M ] [ M ] = [ MQ n ] [ M ] = K [ Q ] n

Rearrange Equation (4) (5) F 0 F = 1 + K [ Q ] n

The quenching data were presented as a plot of F 0 F versus [Q], and the Stern-Volmer plot is an upward curvature, concave towards the y-axis, as shown in Figure 2A.

This result indicates that HSA can be quenched both by collisions and by complex formation with C₆₀ nanoparticles. Then the fractional fluorescence remaining ( F 0 F ) is given by (6) F F 0 = f ( 1 + K SV [ Q ] )where f is the fraction not complexed, and (1 + K_SV[Q]) is the fraction not quenched by collisional encounters.

According to the definition, f can be obtained from Equation (5), and then rearrangement of Equation (6) yields: (7) ( F 0 F − 1 ) = K SV [ Q ] + K [ Q ] n + K SV K [ Q ] n + 1

The nonlinear fitting analysis was performed by plotting ( F 0 F − 1 ) against Q (Figure 2B). From the Stern-Volmer plot, the value of K_SV was 6.56 × 10⁴ L·mol⁻¹. The K_q was easily calculated according to Equation (1), 6.56 × 10¹² L·mol⁻¹·s⁻¹, which is much higher than the maximal collided dynamic quenching constant (2.0 × 10¹⁰ L·mol⁻¹·s⁻¹) [20]. This result indicates that the fluorescence quenching of HSA by the addition of nC₆₀ is mainly caused by static quenching. There is non-linearity obtained from Stern-Volmer when the nC₆₀ concentration is lower than 11.30 μM. When nC₆₀ concentration is higher than 11.30 μM, the intrinsic fluorescence of HSA is significantly quenched. The plot appears to be an upward curvature with increasing nC₆₀ concentration, which is a characteristic feature of mixed quenching. This suggests that it is not a single quenching mechanism that exists in the binding process. The mechanism of HSA quenching caused by water-soluble pristine nC₆₀ is different to that caused by water-soluble nC₆₀ derivative [12].

Synchronous fluorescence spectroscopy is a common method used to provide information about conformational changes of protein since the possible shift of maximum emission wavelength is related to the polarity of the environment. The synchronous fluorescence spectra of HSA-nC₆₀ system are shown in Figure 3. The maximum emission wavelength of tyrosine residues has a small red shift (from 283 nm to 285 nm) when Δλ = 15 nm, indicating that there are some changes in the environment of the tyrosine residues. The maximum emission wavelength of tryptophan residues red shifts from 279.5 nm to 281.5 nm when Δλ = 60 nm (Figure 3B). This suggests that there is a less hydrophobic or more polar environment change around the tyrosine residues and tryptophan residues. This may be ascribed to the fact that the hydrophobic amino acid structure surrounding tryptophan residues in HSA tends to collapse slightly, resulting in tyrosine residues and tryptophan residues being more exposed to the aqueous phase.

Circular dichroism spectra can sensitively monitor conformation changes in the protein upon interaction with the ligand. In this experiment, the calculated α-helicity content of native HSA solution is 54.9%, and with the addition of nC₆₀ to the native HSA solution, the α-helicity content of HSA increased to 59.8%, 61.2% and 62.0% (Figure 4). Apparently, the higher the added nC₆₀ concentration, the more the α-helicity content. The increase of α-helicity content indicated that the binding of HSA and nC₆₀ induces protein folding, which results in some hydrophobic regions becoming more compact.

It is well known that HSA has two major drug binding sites, site I and site II, which are located in the hydrophobic pocket of sub-domain IIA and sub-domain IIIA, respectively. In order to identify the nC₆₀-binding site on HSA, two probes were used. One probe of HSA is dansylamide (DNSA), whose binding site is located in the region of sub-domain IIA (sudlow site I); another probe, dansylproline (DP) is bound to HSA in the sub-domain IIIA (sudlow site II) [21,22]. During the experiment, nC₆₀ was gradually added to the solution of HSA with site probes held in equimolar concentrations (20 μM). No absorption of these two probes to nanoparticles of C₆₀ was observed in capillary electrophoresis analysis (data not shown). As shown in Figure 5, only the probes or HSA alone have no fluorescence signal within the wavelengths 400–600 nm. When the probes and HSA are mixed together, the DNSA-HSA complex has a maximum emission wavelength at 487 nm (excitation wavelength 350 nm) and DP-HSA has a maximum emission wavelength at 480 nm (excitation wavelength 375 nm). With the addition of nC₆₀ to the HSA-probe solution, the fluorescence intensity of the HSA-probe complex was significantly lower than that without nC₆₀, indicating that the binding of DNSA/DP to HSA was affected by adding nC₆₀. The fluorescence intensity of DNSA-HSA was lower than that of DP-HSA, indicating that nC₆₀ could more strongly influence the binding of DNSA to HSA than that of NP to HSA.

Our results above indicate that the fluorescence quenching and conformational changes of HSA can be attributed to the interaction between HSA and nC₆₀. C₆₀ fullerene has also been reported to be bound to the subdomain IIA of HSA using time-resolved fluorescence decay experiments, docking calculations, and binding site alignment methods [13,23]. Here we have employed several approaches in order to address the issue of fullerene interactions with proteins. What needs to be especially pointed out is that it is difficult to incorporate nC₆₀ into the binding pocket because the size of nC₆₀ is larger than protein molecules. The hydrodynamic diameter of the nC₆₀ nanoparticles was determined to be from 50 nm to 110 nm by TEM and DLS, respectively (Figure 6A, B). Ganazzoli et al. reported atomistic computer simulations of some albumin subdomains on a hydrophobic graphite surface [24]. According to these simulations, we proposed a binding model to describe the interaction between HSA and nC₆₀ (Figure 6C). In this model, HSA is adsorbed on the surface of nC₆₀ aggregate which leads to the fluorescence quenching and conformational changes of HSA. The interaction sites were presumed to be near to Site I and Site II of HSA; following the structure and conformation of HSA was affected during adsorption. In our previous study, we have observed that HSA could compete for the surface of nC₆₀ with functional enzymes [25].

C₆₀ (99.5% purity) was obtained from Sigma Chemical Company, St. Louis, USA. HSA (98% purity, essentially fatty acid and globulin free) was purchased from NEB (USA). All solutions were prepared in ultrapure water with the resistivity of 18.2 MΩ·cm (ALGA system, UK). Phosphate buffer (10 mM, pH = 7.4) was prepared with chemicals of analytical pure grade, obtained from commercial sources, and filtered with a 0.22-μmm filter (Millipore, Bedford, MA, USA). All other materials were of analytical pure grade.

C₆₀ suspensions were prepared according to the method described previously [26]. Ten milligrams of C₆₀ was completely dissolved in 10 mL toluene. Then 2.5 mL C₆₀ toluene solution was added into 50 mL water. After ultrasonic treatment for approximately 4 h, toluene was allowed to evaporate under vacuum for at least 1 h. The mean hydrodynamic diameter of the nC₆₀ nanoparticles was determined by transmission electron microscopy (TEM) and dynamic light scattering (DLS) (BI-20OSM laser light scattering spectrometer, Brookhaven, USA).

Absorption spectra were recorded on a UV-2102PCS UV-Vis spectrophotometer (UNICO, USA) equipped with a 1.0 cm quartz cuvette at room temperature. The wavelength range was recorded from 200 to 500 nm.

All fluorescence emission spectra were recorded in the range of 300–450 nm on a LS-55 spectrofluorimeter (PE, USA) equipped with 100 μL quartz cells. The bandwidth of excitation and emission (3 nm) and scan speed (300 nm/min) were kept constant within each data set. Fluorescence intensity at 343 nm was used for calculation in Stern-Volmer equation. All measurements were performed at room temperature (∼25 °C).

Binding site marker competitive experiments were carried out by the fluorescence titration methods. The concentration of HSA and dansylamide/dansylproline (DNSA/DP) were all kept at 10 μM. Then nC₆₀ was gradually titrated into the HSA-DNSA/DP systems. Fluorescence spectra were recorded with an excitation wavelength of 370 nm (for DNSA) and 375 nm (for DNP) in the range of 400–600 nm.

Circular dichroism (CD) spectroscopy was performed on a J-810 spectropolarimeter (Jasco Co., Japan) over a wavelength of 200–260 nm. The scanning speed was set at 500 nm/min. Each spectrum was the average of three successive scans and appropriate buffer solutions running under the same conditions were taken as blank and their contributions were subtracted from the experimental spectra.