A study on human serum albumin corona formed on photoluminescent carbon dots

Characterization of PCDs

The as-prepared PCDs can be readily dissolved in an aqueous solution, and we were first evaluated that of the size and potential by dynamic light scattering (DLS). The zeta-potential result of PCDs is 0.29 ± 0.2 mV (Figure 1(a)) and HSA is −19.3 ± 0.4 mV (Figure 1(c)). The zeta-potential measurement proved that the surface of PCDs is positively charged and HSA is negatively. The average hydrodynamics of PCDs and HSA are 5.17 ± 0.15 nm (Figure 1(b)) and 5.65 ± 0.34 nm (Figure 1(d)), respectively. The particle size of PCDs is small, so they can be used as a probe for detecting biological cells, and the damage to the organism is also minimal.

OPEN IN VIEWER

Figure 1.

The hydrodynamic diameter (D_h) and zeta (ζ) potential of PCDs and HSA. The ζ-potentials of PCDs and HSA are (a) 0.29 ± 0.2 mV and (c) −19.3 ± 0.4 mV. The average D_h of PCDs and HSA are (b) 5.17 ± 0.15 nm and (d) 5.65 ± 0.34 nm, respectively.

With using a UV-Vis spectrophotometer, the aqueous solution of PCDs shows a characteristic absorption spectrum with peaks at 230 and 360 nm, respectively. When PCDs are excited at a wavelength of 360 nm, there is a relatively narrow and symmetric emission spectrum at 455 nm (Figure 2(a)). The emission peak position of the PCDs is unchanged at different excitation wavelengths, and the fluorescence intensity changes with an increase of the excitation wavelength. But there is no red shift, indicating that the particle size of the nanoparticles is relatively uniform. Because of their bright blue appearance under UV illumination and lack of color in daylight, the as-prepared PCDs can act as a fluorescent ink as demonstrated in Figure 2(c) and (d). This feature can be used for anti-counterfeiting and information marking.

OPEN IN VIEWER

Figure 2.

The optical characterization of PCDs. (a) There are two absorption peaks centered at 230 and 360 nm, respectively, and the emission peak is centered at 455 nm using an excitation wavelength of 360 nm. The inset photo shows the sample in daylight (left) and under ultraviolet irradiation (right). (b) The spectra for excitation dependent emissions. The excitation wavelengths change from 280 to 420 nm as shown by the colored lines. Photos of paper written with PCDs under (c) daylight and (d) ultraviolet illumination.

The thermodynamics and kinetics of HSA corona on PCDs

It clearly shows two photoluminescence (PL) bands from both the excitation–emission matrices (EEMs) of HSA (Figure 3(a)) and PCDs (Figure 3(b)). The strongest emission of HSA is centered at 342 nm upon 280 nm excitation, while the as-prepared PCDs show two fluorescent groups, the higher one is at 456 nm with an excitation at 360 nm, and the other emits at 458 nm with an excitation at 252 nm. When excited near 360 nm, the fluorescence intensity of the PCDs is strongest.

OPEN IN VIEWER

Figure 3.

Excitation–emission matrices of (a) HSA and (b) PCDs. The highest PL intensity of HSA is distributed around 342 nm using an excitation at 280 nm. The PCDs show two fluorescent groups, the higher one is at 456 nm with an excitation at 359 nm, and the other emits at 458 nm with an excitation at 252 nm.

From the optical spectra of HSA and PCDs (Figure 4(a)), we can see that the PL intensity of HSA decreased along with an increasing concentration of PCDs, which indicated that HSA PL was quenched by transferring the fluorescence resonance energy to the PCD molecules. The fluorescence quenching method was employed to study both the kinetics and thermodynamics of the physical adsorption procedure. From Figure 4, it can be seen that the fluorescence of HSA was quenched more and more by increasing the addition of PCDs. The Stern–Volmer quenching constant (which is a measure of the quenching efficiency) was calculated as 1.02 × 10⁶ M⁻¹. Therefore, a modified Stern–Volmer equation should be used to analyze the quenching data.

OPEN IN VIEWER

Figure 4.

PCD quenching HSA PL assay at 298 K. (a) PL spectra of HSA (6.0 μM) on mixing with various amounts of PCDs and excitation at 280 nm. From top to bottom at a wavelength of 342 nm, the final concentrations of PCDs are 0, 10, 20, 30, 40, 50, 60, 70, 80, and 90 μg mL⁻¹. (b) The Stern–Volmer plot (black dots) and the corresponding linear fit (red line) for the quenching data from (a). (c) The modified Stern–Volmer plot (black dots) and the corresponding linear fit (red line) for the quenching data from (a). (d) The Hill plot (black dots) and the corresponding linear fit (red line) for the quenching data from (a). The error bars were obtained from three replicates.

In this case, the as-prepared PCDs and HSA form a protein corona, and hence K_SV can be considered as K_a. Therefore, we employed the modified Stern–Volmer equation to further analyze the quenching data at three different temperatures (298, 304, 310 K; see Supplemental Figures S1 and S2 in Supporting Information). The resulting adsorption data at three different temperatures are shown in Table 1, and it is clear that K_a correlates positively with temperature, which further confirms that the HSA–PCDs adsorption is initiated by an electrostatic interaction (dynamic quenching). In addition, the value of the thermodynamic parameters in the protein reaction can be used to identify primary forces that contribute to protein and ligand stability. The thermodynamic changes of the reaction system are generally judged according to ΔH₀, ΔS₀, and ΔG₀. According to the van’t Hoff equation and the Gibbs–Helmholtz equation, the ΔG₀ values at three temperatures (298, 304, and 310 K) are all negative. Negative values indicate that the incorporation between HSA and PCDs is spontaneous, while a negative value of ΔH₀ and a positive value of ΔS₀ also indicates that the reaction is spontaneous. When ΔH₀ is less than zero, the physical adsorption process between HSA and PCDs will release heat to the surrounding environment. The reason why ΔS₀ is greater than zero is that the entropy of the surrounding environment increases after the endothermic reaction, and the increase of environmental entropy exceeds the decrease of the entropy of the HSA–PCDs system. Therefore, the entropy of the system plus the entropy of the environment is still greater than zero, fully in line with the second law of thermodynamics.

OPEN IN VIEWER

Table 1.

Significant parameters obtained from the modified Stern–Volmer formula at different temperatures.

Temperature (K)	Ksv (M−1)	Ka (M−1)	n	ΔH0 (kJ mol−1)	ΔS0 (J mol−1 K−1)	ΔG0 (kJ mol−1)
298	1.02 × 106	9.12 × 106	1.85	−22.60	57.85	−39.84
304	1.12 × 106	9.61 × 106	1.93			−40.19
310	1.13 × 106	1.16 × 107	2.01			−40.53

Demonstration of biolabelling applications

Due to the fluorescence properties and biocompatibility of PCDs, they can serve as be good substitutes for traditional quantum dots for bio-medical applications such as in labeling, imaging, sensing and even photothermal-based therapy. Here we selected bean sprouts as a typical and facile biological sample model in order to demonstrate the labeling and imaging feasibility of the as-prepared PCDs. It is clear from Figure 5 that bean sprouts show bright blue fluorescence under UV illumination, indicating the homogeneous labeling of PCDs. This demonstration set a successful application example for our lab-made PCDs, which hold great potential for other biomedical uses.

OPEN IN VIEWER

Figure 5.

Demonstration of PCDs for labeling bean sprouts: (a) a photo of bean sprouts labeled with PCDs under daylight and (b) a photo of PCD-labeled bean sprouts taken under UV illumination.

Chemicals and materials

Citric acid, ethylenediamine, HSA, and quinine sulfate (purity 98%, suitable for fluorescence quantum yield measurements) were purchased from Sigma-Aldrich (New York, USA). Tris-HCl (analytical reagent grade) was purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). All chemicals and reagents were used directly without further purification. Bean sprouts were purchased from a local supermarket in city.

PCD synthesis

Citric acid (1.0 g) and ethylenediamine (384 μL) were added to Milli-Q water (10 mL) to prepare a colorless reaction solution, which was then heated in a microwave oven (750 W) for 4–5 min, during which time the solution changed from colorless to brown indicating the formation of PCDs. ³³ The aqueous solution was subjected to dialysis against Milli-Q water with a cellulose ester membrane bag (molecular-weight cutoff: 1 kDa) to remove any unreacted chemicals and small molecules, and then the solution was subjected to centrifugation at 10,000 r min⁻¹ for 10 min. The obtained pellet was dried under vacuum for 48 h and redispersed in Milli-Q water to a concentration of 0.1 mg mL⁻¹.

To test the synthesis conditions, two laboratory microwave ovens (1. MCR-3, Tianjin Yuhua Instrument Co., Ltd., China; 2. MCR-3, Shanghai Ledun Instrument Co., Ltd., China) were compared with one domestic microwave oven (WD750ASL23, Guangdong, China). Using the same electric power (all set to 750 W) and heating for 4-5 min, no significant difference between the resulting solutions (from both spectrometer and DLS characterizations) between the laboratory and domestic microwave ovens was observed. However, both the power and heating time can produce huge differences for the synthesis of PCDs, and the power of 750 W and heating time of 4-5 min were the optimized conditions for the current synthesis.

Dynamic light scattering

The hydrodynamic diameter (D_h) and zeta potential of PCDs were measured using commercial DLS equipment (Malvern Zetasizer Nano-ZS90; Malvern Instruments Ltd., Worcestershire, UK) equipped with quartz or disposable cuvettes (DTS0012, minimum sample volume is 1.0 mL) and fitted with a 633-nm He-Ne laser beam. Measurements were carried out at 25°C and at a 90° scattering angle. The D_h and ζ-potential were obtained by cumulative analysis (Zetasizer software Nano-ZS90; Malvern Instruments Ltd.).

Spectroscopy

The absorbance and the PL spectra were recorded on a UV-Vis spectrometer (U-2900; Hitachi, Tokyo, Japan) and a fluorescence spectrometer (Fluorolog^®-MAX 4; Horiba, Kyoto, Japan), respectively. Quartz cuvettes (light path is 1.0 cm) were used for all spectral measurements. For it fluorescence measurements, both the excitation and emission slits were set at up to 5 nm.

Quantum yield

The quantum yield ( φ ) of the as-prepared PCDs was determined by a comparative method. In brief, a standard solution of quinine sulfate (reported quantum yield: 54%) was prepared by dissolving the substance in 0.1 M H₂SO₄ (refractive index η = 1 . 33 ). Aqueous solutions containing varying concentrations of PCDs were prepared using Milli-Q water ( η = 1 . 33 ) as the solvent. The absorbancies of the solutions at the excitation wavelength were recorded using a UV-Vis spectrophotometer. PL emission spectra of all the samples were recorded at an excitation wavelength of 360 nm. The samples were then analyzed by comparing the integrated PL emission intensity at the excitation wavelength where PCDs and the reference showed the same absorbance. The quantum yield ( φ x ) was calculated using the following equation (equation (1))

where φ s t is the quantum yield of the standard sample (quinine), Grad is the gradient/slope of the plot of integrated fluorescence intensities against absorbance at a certain temperature. The subscripts x and st are the sample weighting for detection and the standard sample. η is the refractive index of the solvents.

Fluorescence quenching

This method refers to previous publications with minor modifications.^{27–29,34–35} In brief, the excitation was fixed at 280 nm and the slits for both excitation and emission were set at 5.0 nm in all PL measurements. The temperature was controlled by a water bath and the samples were incubated for at least 30 min. For all the calculations, all the fluorescence intensities were corrected using the following equation (to remove the inner filter effect, equation (2))

Both the Stern–Volmer equation (equation (3)) and its modified formula (equation (4)) and the Hill model (equation (5)) were employed to calculate the binding constants and sites

where F_c and F_m are the corrected and measured PL intensity, and F₀, F, and F_sat are the PL intensities in the absence, presence, and saturated state of quencher (PCDs), respectively. Furthermore, A_ex and A_em are the absorption values of the system at the excitation and emission wavelengths, k_q is the quenching constant, τ₀ is the lifetime of the fluorophore in the absence of quencher, and [Q] is the concentration of the quencher. f_a is the fraction of accessible fluorophores, and K_a is the effective quenching constant for the accessible fluorophores. The binding sites can be obtained by n.