Domain-Specific Association of a Phenanthrene–Pyrene-Based Synthetic Fluorescent Probe with Bovine Serum Albumin: Spectroscopic and Molecular Docking Analysis

Abstract

In this report, the interaction between a phenanthrene–pyrene-based fluorescent probe (PPI) and bovine serum albumin (BSA), a transport protein, has been explored by steady-state emission spectroscopy, fluorescence anisotropy, far-ultraviolet circular dichroism (CD), time-resolved spectral measurements, and molecular docking simulation study. The blue shift along with emission enhancement indicates the interaction between PPI and BSA. The binding of the probe causes quenching of BSA fluorescence through both static and dynamic quenching mechanisms, revealing a 1:1 interaction, as delineated from Benesi–Hildebrand plot, with a binding constant of ∼10⁵ M^–1, which is in excellent agreement with the binding constant extracted from fluorescence anisotropy measurements. The thermodynamic parameters, ΔH°, ΔS°, and ΔG°, as determined from van’t Hoff relationship indicate the predominance of van der Waals/extensive hydrogen-bonding interactions for the binding phenomenon. The molecular docking and site-selective binding studies reveal the predominant binding of PPI in subdomain IIA of BSA. From the fluorescence resonance energy transfer study, the average distance between tryptophan 213 of the BSA donor and the PPI acceptor is found to be 3.04 nm. CD study demonstrates the reduction of α-helical content of BSA protein on binding with PPI, clearly indicating the change of conformation of BSA.

Introduction

The interaction and energetics of protein binding toward small molecules are largely dependent on the microenvironment and molecular architecture arising due to folding/unfolding or even change of the protein structure. The remarkable properties of a small molecule in such a microenvironment bear information related to the binding site, which is essential for drug development and many other investigations.¹⁻⁴

Model globular proteins, such as serum albumins, are important transport proteins and are found plentiful in plasma.⁵⁻⁸ Bovine serum albumin (BSA), a large globular protein (65 000 Da), contains 583 amino acid residues in a single chain.⁹ The three domains with different surface charge densities impact BSA adsorption on charged surfaces.^10,11 As for example, the presence of both positively charged residues (lysine and arginine) and negatively charged amino acids (glutamic acid and aspartic acid) on BSA can result in electrostatic interactions with both negatively and positively charged surfaces, respectively.^12,13 Because of the presence of a negatively charged domain, BSA is involved in (a) binding with water, salts, fatty acids, vitamins, and hormones and carries them between tissues and cells, (b) removing toxic substances, including pyrogens, from the medium, (c) solubilizing lipids and is a blocking agent in western blot or enzyme-linked immunosorbent assay applications, and (d) solubilizing other proteins (e.g., labile enzymes). BSA is readily soluble in water and can only be precipitated in the presence of high concentrations of neutral salts such as ammonium sulfate. However, albumin is readily coagulated by heat. So, it is apparent that the BSA can bind a large variety of bioactive molecules by various noncovalent interactions such as hydrophobic, hydrophilic, and ionic interactions. Tryptophan (Trp) 134 and Trp 213 are the two Trp residues present in BSA. It has three domains I, II, and III, each consisting of two subdomains A and B.

The major binding sites of BSA are localized in subdomains, IIA and IIIA, known as site I and site II.^14,15 To infer the protein interaction site with small molecules, site marker fluorescent probes are generally utilized. A great deal of research activities on the structure and function of serum albumins are reported in the literature.¹⁶⁻²⁷ Nowadays, it is of interest to develop and use special polarity-sensitive fluorescent probes. The main focus of the present work deals with the fluorescence emission and binding aspects of a synthetic fluorescent probe, a phenanthrene–pyrene based conjugate (PPI), in the hydrophobic milieu of a globular protein, BSA, under physiological conditions.

The novelty of the present study stems from the fact that the phenanthrene imidazole molecules²⁸ act as potential selective biomarkers for inhibition of various enzymatic processes. Here, pyrene moiety has been incorporated with phenanthrene imidazole core as a fluorophore unit to validate the binding proficiency with BSA. The presence of an aryl-heteroatom bond, particularly the C–N bond, is significant for showing different biological activities.²⁹

Results and Discussion

PPI was synthesized from phenanthrene-9,10-dione, 1-pyrene carboxaldehyde, and ammonium acetate in glacial acetic acid, as shown in Scheme 1, and characterized by proton nuclear magnetic resonance (¹H NMR) (Figure S1), ¹³C NMR (Figure S2), high-resolution mass spectrometry (HRMS) (Figure S3), and infrared (Figure S4) spectroscopy.

Scheme 1. Synthesis of PPI.

Ultraviolet–Visible (UV–vis) Absorption Study

Absorption spectral study is a useful tool to explore the structural variations and to analyze the complex formation between the protein and probe in solution.³⁰ UV–vis titrations were carried out at 5 μM BSA concentration, gradually increasing the concentration of PPI (0–25 μM) in aqueous buffer solution. The absorption spectral changes of BSA with a gradual change in the concentration of PPI are shown in Figure 1. Though PPI has no absorption at 280 nm, gradual addition of PPI results in an increase in absorbance of BSA at 280 nm, with contemporary growing of an absorbance peak at 380 nm due to PPI. The pattern of the absorption spectra at 280 nm also changes with the increase in the concentration of PPI. This observation supports the complexation between PPI and BSA.

Figure 1.

(A) Absorption spectral changes of BSA (5 μM) with incremental addition of PPI (0–25 μM) at 25 °C. (B) Inset: absorbance plot at 280 and 380 nm as a function of PPI concentration.

Fluorescence Emission Study

Interaction between the protein and probe is well-characterized by the investigation of steady-state fluorescence emission technique. To follow the PPI—BSA interaction, fluorescence titrations were performed at 20 μM PPI concentration in aqueous medium with the incremental addition of BSA. The emission maximum of PPI was shifted from 476 to 457 nm in 70 μM BSA solution with progressive enhancement of the fluorescence intensity (Figure 2A) when PPI was excited at 380 nm. The fluorescence intensity variations and λ_em^max of PPI as a function of the BSA concentration are more clearly exhibited in Figure 2B, which represents a steep variation of λ_em up to 15 μM BSA, followed by attainment of a tableland region, and this fact is a clear indication for an ample modification of the surrounding of PPI within the protein heterogeneous microenvironment.

Figure 2.

(A) Emission spectra of PPI (20 μM) with the gradual addition of BSA (0–70 μM) at 25 °C. (B) Plot of relative variation (F/F₀) of the emission intensity and emission maximum (λ_em^max) of PPI against BSA concentration.

This noticeable blue shift with a concomitant increase in the fluorescence intensity is caused by the alteration in the position of PPI from a more polar aqueous phase to a more hydrophobic protein environment when PPI binds with BSA. BSA fluorescence comes from the presence of three amino acid residues namely, tyrosine (Tyr), Trp, and phenylalanine. In particular, Trp fluorescence is used to monitor the changes in the structural conformation of the BSA protein and to interpret the local environment of BSA-bound PPI.³¹⁻³³ There are two Trp moieties in BSA, that is, Trp 134 and Trp 213, which are located in subdomains IB and IIA, respectively. Trp 134 is well-exposed to the hydrophilic region, whereas Trp 213 resides in a hydrophobic cavity of the BSA protein.^34,35

BSA shows a strong fluorescence maximum at 342 nm in aqueous buffer solution when the excitation of BSA is made at 295 nm.³⁶ Excitation at 295 nm was selected to minimize the contribution of the Tyr residue present in BSA. To find out the binding constant for BSA–PPI interaction, in another experiment, the fluorescence titration was performed at 10 μM BSA concentration with the gradual addition of PPI leading to saturation. Figure 3A shows that the emission intensity of BSA is decreased considerably along with a small blue shift of λ_em^max from 342 to 338 nm with increasing PPI concentration, which in turn implies that PPI binds strongly with BSA and also indicates that the microenvironment around Trp moieties present in BSA is modified on interacting with PPI.³⁷ There is an important observation that apart from the quenching of emission intensity of BSA at 342 nm, a new emission band is developed at 461 nm. The intensity of this emission band is progressively enhanced with the increasing concentration of PPI. The presence of an isoemissive point at 420 nm is an indication of the equilibrium between the free and bound forms of PPI. Inset of Figure 3A vividly shows the emission intensity variation at 342 and 461 nm as a function of PPI concentration.

Figure 3.

(A) Emission spectra of BSA (10 μM) with the gradual addition of PPI (0–20 μM) at 25 °C. (B) Inset of Figure 3A; shows the plot of emission intensity variation at 342 and 461 nm, respectively, as a function of the PPI concentration. (C) Representative Benesi–Hildebrand plot for 1:1 complexation of BSA with PPI. λ_ex for BSA is 295 nm.

Probe-Protein Binding Study

The binding of PPI with BSA can be explained by eqs 1 and 2, considering 1:1 complexation between them.

1

2

where K_BH represents the association constant. The data obtained from the spectrofluorimetric titration of a fixed concentration of BSA with the increasing concentration of PPI are further investigated to find out the binding constant by adopting Benesi–Hildebrand equation.³⁸

3

where ΔF_max = |F_∞ – F₀| and ΔF = |F_x – F₀|. F₀, F_x, and F_∞ indicate the emission intensities of BSA in the native state, at an intermediate PPI concentration, and at a PPI concentration when the interaction is saturated, respectively. |F_∞ – F₀|/|F_x – F₀| versus 1/[PPI] plot displays a linear variation (Figure 3C), validating the accuracy of eq 3, and supports the 1:1 complexation between PPI and BSA. The 1:1 binding of PPI with BSA was also confirmed by Job’s plot analysis using the emission spectral data. In this method, the emission data were recorded by changing the PPI:BSA molar ratio, whereas the total molar concentration of PPI and BSA was constant.³⁹ The Job’s plot for PPI–BSA, that is, the difference in the emission intensity at 342 nm versus the mole fraction of PPI (Figure S5), intersected at 0.491, showing the number of PPI molecules binding to BSA to be around unity. The K_BH value, evaluated from the reciprocal of the slope 1/K_BH, and the corresponding free energy change (ΔG) accompanying the binding process are presented in Table 1, implying strong complexation between PPI and BSA.⁴⁰

Table 1. Binding Parameters for the Association of PPI with BSA at 25 °C.

method	environment	binding constant (105 M–1)	ΔG (KJ mol–1)
Benesi–Hildebrand	PPI–BSA	(1.12 ± 0.06)	–28.80
fluorescence anisotropy	PPI–BSA	(1.55 ± 0.25)	–29.60

Fluorescence Anisotropy Study

To gather information about the rigidity of the surrounding environment of a probe, fluorescence anisotropy measurement is a useful experiment.⁴¹ It also gives information regarding the boundary to which the rigid environment obstructs the rotational mobility of the probe. Enhancement of anisotropy values reflects an increase in the rigidity of the environment around a fluorescent probe. The change in the anisotropy values of PPI with a change in the BSA concentration in aqueous buffer medium is represented in Figure 4A. Primarily, a rapid enhancement of the anisotropy value from 0.067 to 0.235, till the addition of 8 μM BSA, was observed; then, the increase was gradual to a value of 0.284 till the addition of 40 μM BSA. The increasing value of anisotropy clearly indicates the fact that substantial restriction is imposed on the free motion of PPI molecules with the gradual addition of BSA, and this can be only possible if BSA strongly binds with PPI. The maximum anisotropy value, 0.293, was obtained at 60 μM BSA, and after that, the levelling off of the anisotropy values was observed, which in turn reflects the saturation of association between PPI and BSA.

Figure 4.

(A) Anisotropy variation of PPI as a function of the BSA concentration at 25 °C. λ_ex = 380 nm and λ_em = 475 nm for PPI. (B) Plot of 1/f_B vs 1/[BSA] for evaluating the binding constant of the PPI–BSA composite from the anisotropy data.

In accordance with Ingersoll and Strollo⁴² method, the binding constant of the PPI–BSA composite can be ascertained by adopting the following equation

4

where K_b denotes the apparent binding constant of the PPI–BSA composite. f_B corresponds to the fractional fluorescence contribution of PPI bound to BSA, as shown in eq 5.

5

where the anisotropy values of bound BSA–PPI and free PPI are denoted by r_B and r_F, respectively. The correction factor, R, that is, the ratio of I_B and I_F, is taken into consideration to confirm the fact that PPI experiences fluorescence intensity variation on binding with BSA. The double reciprocal plot of 1/f_B versus 1/[BSA] (Figure 4B) is a straight line, and from the slope, the calculated value of K_b is presented in Table 1, which is in good accordance with the value obtained from the spectrofluorimetric titration experiment. Thus, this method sets up its practical application and feasibility to find out the binding constant.³⁷

8-Anilino-1-naphthalene Sulfonic Acid (ANS) Displacement Assay

To check the possible binding site of PPI on BSA, the ANS displacement assay was carried out. The fluorescent probe ANS was used to get information about the hydrophobic binding locations of the protein.³⁹ In accordance with the procedure, the displacement studies were accomplished with the introduction of ANS, maintaining similar conditions. Figure S6 represents the plot of F/F₀ versus PPI concentration, which clearly shows that at 20 μM concentration, PPI has a better quenching influence on the emission intensity of BSA than ANS, that is, PPI could quench ∼68% and ANS around 51%. The emission intensity of ANS was considerably augmented at 470 nm upon interaction with the hydrophobic regions of BSA, but when PPI was added to the BSA–ANS composite (1:1), the emission intensity of the composite decreased around 60%. This observation suggested that PPI moderately contends with ANS for the hydrophobic locations of BSA by removing the bound ANS molecules, leading to a decrement in the emission intensity of the BSA–ANS composite.

Site-Selective Binding of PPI on BSA

The competitive fluorescence displacement studies were executed to ascertain the BSA binding site in which PPI is located, using two well-known drugs (warfarin and ibuprofen). The range of the binding interaction of the PPI–BSA composite can also be revealed by observing the fluorescence intensity variation of the system.³⁷ Site marker warfarin exclusively binds at subdomain IIA of site I by hydrophobic interaction, whereas ibuprofen precisely binds at subdomain IIIA of site II through hydrophobic, hydrogen-bonding, and electrostatic interactions.^15,37,43,44 Here, PPI was progressively added to the BSA–site marker composites (1:1) in aqueous buffer solution to reveal the spectral change with PPI. Figure 5 represents the spectral changes influenced by the presence of site markers.

Figure 5.

Effect of site markers on the PPI–BSA composite. (A) [BSA] = [Warfarin] = 10 μM with the addition of PPI, each time 2 μM, to a total concentration of 12 μM from curves 3–8 at 25 °C. (B) [BSA] = [Ibuprofen] = 10 μM with the addition of PPI, each time 2 μM, to a total concentration of 14 μM from curves 3–9 at 25 °C. λ_ex of BSA = 295 nm.

The introduction of warfarin site marker into the BSA solution significantly quenched the fluorescence intensity associated with a red shift of λ_em from 342 to 358 nm (Figure 5A). Then, incremental addition of PPI into the BSA–warfarin composite results in a gradual decrease in the fluorescence intensity, which in turn indicates the PPI influence on the binding of warfarin to BSA. In contrast to warfarin, no meaningful change of BSA fluorescence intensity was observed upon addition of ibuprofen (Figure 5B). PPI influences the quenching of fluorescence intensity of the BSA-ibuprofen composite nearly to the same extent as in the absence of ibuprofen (Figure 3A). Therefore, the above experimental studies and outcomes clearly establish the fact that the binding of PPI to BSA is principally located at subdomain IIA of site I, which indicates that Trp 213 is inside or in the vicinity of the PPI binding site.

Study of Fluorescence Quenching Induced by PPI

The incremental addition of PPI into BSA solution induces the quenching of BSA fluorescence (Figure 3A). The fluorescence quenching mechanism can be static, owing to the ground-state association between the fluorophore and the quencher, or dynamic because of collisional encounters between the above said two species at the excited state. These two quenching mechanisms can be discriminated by studying the lifetime measurements or by their varying dependence on temperature and viscosity.⁴⁵ We have performed the emission quenching of Trp present in BSA with the incremental addition of PPI in aqueous buffer solution. Because PPI has a considerable absorbance at ∼280 nm and to avoid the involvement of the inner filter effect to the quenching of BSA fluorescence, the emission intensity was corrected by taking the following relation^39,46

6

where, F and F_obs represent the corrected and observed emission intensities, respectively, of the sample under study. A_ex and A_em denote the absorbance value at the excitation and emission wavelengths, respectively. Figure S7 represents the emission spectrum of PPI at λ_ex = 295 nm. The probable fluorescence quenching mechanism of BSA–PPI complexation was verified by analyzing the emission data using the well-known Stern–Volmer equation⁴⁵

7

where, F₀ and F correspond to the BSA emission intensities in free form and with the successive addition of PPI, respectively. K_SV represents the Stern–Volmer constant, and k_q is the bimolecular quenching rate constant. [PPI] and ⟨τ₀⟩ are the molar concentration of the quencher and the average lifetime of a BSA molecule in the absence of PPI, respectively. The involvement of only one type of quenching mechanism, that is, either static or dynamic, is inferred by the linear Stern–Volmer plot. The occurrence of both the abovementioned quenchings can be inferred, when the plot displays an upward deviation.^{32,45,47−49}Figure 6A displays an upward curvature, signifying the coexistence of static and dynamic quenchings with the same quencher (PPI), and/or the extent of quenching is high at a higher concentration of PPI. Here, F₀/F is linked with [PPI] by the modified form of Stern–Volmer equation⁴⁵

8

9

where dynamic and static quenching constants are represented by K_D and K_S, respectively. The first factor of the right-hand side in eq 8 represents dynamic quenching, whereas the second factor represents static quenching. The presence of [PPI]² term in eq 9 accounts for the observation of an upward deviation at high [PPI] when both of the above said quenchings take place for the same quencher. The observed dynamic portion can also be ascertained by fluorescence lifetime measurements of BSA against PPI concentration using the following equation

10

where ⟨τ₀⟩ and ⟨τ⟩ correspond to the average lifetime of BSA in native state and with the incremental addition of PPI, respectively. The value of K_D = (9.891 ± 0.4) × 10³ M^–1 is obtained from the slope of the plot ⟨τ₀⟩/⟨τ⟩ versus [PPI] (Figure 6B). The value of K_q = K_D/⟨τ₀⟩ = (1.64 ± 0.06) × 10¹² M^–1 s^–1 is obtained by using the value of K_D and ⟨τ₀⟩ =6.01 ns. The calculated K_q value is 2 orders of magnitude greater than the maximum diffusion-controlled K_q value, 2.0 × 10¹⁰ M^–1 s^–1.⁴⁵ This implies that the quenching of BSA fluorescence by PPI occurs through Coulombic resonance interaction but not operated by the diffusion process.³² The fluorescence emission data were further studied by applying the modified Stern–Volmer equation^45,46,50

11

where f is the maximum accessible fractional initial fluorescence of the protein molecule to the quencher. K_SV and f values are determined from the intercept and slope of the plot F₀/(F₀ – F) versus 1/[PPI] (Figure S8). The evaluated value of K_SV is (5.926 ± 0.007) × 10⁴ M^–1. The calculated f value 1.54 indicates that 64.65% of BSA fluorescence is accessible to PPI. The value of K_q is found to be (9.73 ± 0.01) × 10¹² M^–1 s^–1, and this high value again establishes the fact that the quenching of BSA is not operated by the diffusion-controlled process.

Figure 6.

(A) Representative Stern–Volmer plot (from steady-state fluorescence study) for the quenching of BSA (10 μM) fluorescence by PPI at 298 K (λ_ex = 295 nm and λ_em = 342 nm). (B) Time-resolved Stern–Volmer plot for the quenching of BSA (10 μM) fluorescence by PPI at 298 K.

Analysis of BSA–PPI Binding Equilibria and Determination of Thermodynamic Parameters

The results obtained from the above dynamic interaction have been utilized to isolate static and dynamic quenchings in eq 8. Static quenching is defined by the following equation

12

where n signifies the stoichiometry of the binding process, that is, the number of PPI molecules associated with each BSA molecule. From Figure 6A, it is clearly observed that the contribution of dynamic quenching, that is, the upward curvature, becomes considerable only for [PPI] > 8 μM (molar ratio [PPI]/[BSA] > 0.8). So, at a PPI concentration below 8 μM, static quenching can be considered exclusively, as evidenced from a linear dependence of F/F₀ on [PPI]. The trend of linear Stern–Volmer quenching plots for [PPI] < 8 μM at different temperatures is exhibited in Figure 7A. By varying the temperature from 288 to 308 K, the decreasing tendency of the quenching plot is observed, and the upward deviation becomes insignificant. On the basis of such type of quenching plot and data, the value of the binding constant (K_b) and the value of n (number of binding sites) can be calculated by using eq 13(45,51,52)

13

Figure 7.

(A) Representative linear Stern–Volmer plots of PPI-induced quenching of BSA (10 μM) fluorescence at low PPI concentrations at different temperatures. (B) Double log plots for the determination of the number of binding sites and the binding constant value of PPI–BSA complexation at different temperatures.

Figure 7B exhibits the representative log[(F₀ – F)/F] versus log[PPI] plot at different temperatures, and all numerical parameters thereby obtained are tabulated in Table S1, which shows that the K_b value decreases with increasing temperature. Therefore, it can be expected that static interaction is a temperature-dependent process. The high value of K_b implies a strong binding affinity of PPI to BSA. Basically, four types of noncovalent forces, namely, van der Waal forces, multiple hydrogen-bonding, hydrophobic, and electrostatic interactions play a vital role in the binding of probes with proteins.^49,53 To ascertain the nature of the ground-state interaction between PPI and BSA in terms of the aforesaid noncovalent forces relevant with the complexation process, related thermodynamic parameters have been estimated by using the following van’t Hoff equations⁵⁴

14

15

where ΔH°, ΔS°, and ΔG° are the standard enthalpy, entropy, and free energy changes, respectively, for the binding process. R represents the molar gas constant. The values of ΔH° and ΔS° are evaluated from the slope and intercept of the ln K_b versus 1/T plot (Figure 8). ΔG° value is then calculated by using eq 15. The values of all related thermodynamic parameters are listed in Table S1, which show that the value of ΔH° is highly negative and ΔS° also carries a negative value. This implies that the association of PPI with BSA results from the primary contribution of van der Waals interactions, followed by the involvement of extensive hydrogen bonding interactions between PPI and BSA.⁵⁵ Here also, a net negative value of ΔG° implies that the interaction between PPI and BSA is spontaneous and thermodynamically favorable.

Figure 8.

van’t Hoff plot for the interaction of PPI with BSA at different temperatures.

Fluorescence Lifetime Studies

Time-resolved fluorescence decay studies were further accomplished to explore the local microenvironment surrounding the excited probe in the proteinous environment.^17,56 To investigate the dynamics of PPI within the proteinous environment, a nanosecond lifetime decay study of PPI was performed in the absence and with the incremental addition of BSA. Representative decay profiles are shown in Figure S9A. The values of all related parameters are tabulated in Table S2. The lifetime decay profiles were fitted with a biexponential form instead of a monoexponential form, unless the decay curve did not fit well with the monoexponential form. The best fit for the decay profiles was carried out with acceptable values of χ² (χ² within 1.0–1.1). The probe PPI in an aqueous buffer solution is found to display a monoexponential decay with a lifetime of 4.18 ns. Table S2 reveals that the decay profile is changed from the mono- to biexponential form with two lifetime values in the presence of BSA. This is an indication toward separation of PPI into two different environments upon interaction with BSA. Here, we choose to use the average lifetime value in place of more emphasis on the individual fluorescence decay component in such a biexponential form. Table S2 shows that the average lifetime (⟨τ⟩) of PPI gradually decreases with the incremental addition of BSA. The existence of binding interaction is indicated by the meaningful difference in the lifetime values between free PPI and the PPI–BSA composite.

The lifetime decay study was also performed to explore the quenching mechanism (i.e., whether it is static or dynamic or both) of BSA fluorescence by PPI. Here, lifetime decay study of BSA was executed in the absence and with the successive addition of PPI. Representative decay profiles are shown in Figure S9B. The values of all related parameters are incorporated in Table S3. The native BSA displays a biexponential decay profile in an aqueous medium, with ⟨τ⟩ value of 6.01 ns having two decay time components of 3.69 and 6.68 ns with the corresponding relative amplitudes of 22.12 and 77.78%, respectively. This biexponential decay pattern of the native BSA has been reported earlier and attributed to the existence of two Trp moieties at distinct conformational states in two different local environments.^57,58 Fleming and co-workers⁵⁹ have examined the lifetime decays of Trp by considering a model, which is based on the conformational rotamers around the C^α–C^β bond and the comparative charge transfer rate from indole to many electrophiles. The three conformational rotamers of Trp are portrayed below:

Here, the rotamer (C) signifies the faster component, and on the contrary, the relatively slower component generally appears from quick interconversion of (A) and (B) rotamers. Nevertheless, the alteration of the relatively stable (C) rotamer to either A or B form is quite impossible on a nanosecond time scale.⁶⁰⁻⁶² Furthermore, it is supposed that the puckered conformation of the indole ring in the ground state turns into a planar form upon photoexcitation probably due to the delocalization of lone pairs on nitrogen, including the aromatic system. Generally, the distortion of indole ring planarity is due to the interaction with the quencher, implying the alteration of the microenvironment in the vicinity of Trp, which is the primary cause for a decrease in the lifetime value.⁶⁰⁻⁶² A quick look at Table S3 reveals that the relative contribution (α₂) of the faster decay time component (τ₂) progressively decreases from 77.88 to 69.48%, and at the same time, the relative contribution (α₁) of the slower decay time component (τ₁) gradually increases from 22.12 to 31.52%, with the increasing concentration of PPI. Table S3 also shows that the two decay components τ₁ and τ₂ are gradually lowered with the increasing concentration of PPI than the respective values for free BSA. The ⟨τ⟩ value of BSA gradually decreases from 6.01 ns in aqueous buffer solution to 5.01 ns with the gradual addition of PPI (inset of Figure S9B). The decrease in the ⟨τ⟩ value is a clear outcome of substantial interactions between BSA and PPI. To investigate the occurrence of dynamic quenching, a time-resolved Stern–Volmer plot was made by using the value of average lifetime, and it can be defined by eq 10. Figure 6B represents the time-resolved Stern–Volmer plot of BSA bound PPI, and it increases linearly with the PPI concentration, which in turn implies the occurrence of dynamic quenching of BSA fluorescence.

Time-Resolved Anisotropy Decay

The study of lifetime anisotropy decay is useful for garnering knowledge about the rotational motion and relaxation of a fluorescent probe within the proteinous environment.⁶³ To acquire more information about the neighboring microenvironment of PPI, anisotropy decay study of PPI was performed in aqueous medium in the absence and presence of the BSA protein. The representative anisotropy decay profiles are exhibited in Figure 9. PPI exhibits a monoexponential decay profile with a reorientation time of ∼494 ps, indicating a homogeneous environment around PPI. But interestingly, the anisotropy decay profile of PPI is greatly altered in the presence of BSA leading to a dip-and-rise pattern. This type of pattern signifies the co-occurrence of at least two classes of PPI populations, one with a slower rotational correlation time (τ_1r) 6.78 ns having a component (α_1r) of 88% and another with a faster rotational correlation time (τ_2r) 0.79 ns having a component (α_2r) of 12%. In accordance with the reported literature⁶⁴⁻⁶⁶ on the explanation of such type of dip-and-rise pattern, the faster motion is ascribed to the existence of solvent-exposed groups or moieties of the probe, whereas, comparatively slower motion corresponds to the bound counterpart. Another probable explanation can be understood in relation with the rotational diffusion of the probe bound to two discrete binding sites (i.e., hydrophilic and hydrophobic zones) in BSA.^67,68 The study of molecular docking (under the Section of Molecular Docking Results) shows in support of plausible location of PPI to be in the hydrophobic binding region (i.e., in subdomain IIA of BSA). Actually, a considerable population in the hydrophilic region (i.e., in subdomain IB of BSA) seems physically not sound for neutral PPI. The above statement is only valid when the components (i.e., α_1r and α_2r) reveal the relative PPI population in the two interaction sites. The abovementioned data obviously show that the finding probability of PPI in one binding site is appreciably higher than in the other (α_1r > α_2r).⁶⁷⁻⁷¹ So, the observed dip-and-rise pattern can therefore be explained by considering the fact that the probe experiences different types of rotational motions in the protein environment.⁶⁹⁻⁷¹ Such type of anisotropy decay pattern has been illustrated by the related exponential model, which links the decay parameters with the discrete anisotropy parameters as follows⁶⁴⁻⁶⁶

16

where

17

and

18

where, the ith rotational correlation time is indicated by θ_i. α_i stands for the amplitude of the ith lifetime decay component (i.e., τ_i). r(0) denotes the limiting anisotropy. Generally, this type of anisotropy decay pattern has been described⁶⁴⁻⁶⁶ from the outcome of the co-occurrence of two distinctly different lifetime values, which validate the importance of the time-dependent weighing factor f_i(t) in narrating such anisotropy profile, as demonstrated in eqs 16 and 17.

Figure 9.

Lifetime anisotropy decay of PPI (λ_ex = 370 nm and λ_monitored = λ_em) in the (A) absence of BSA and (B) presence of BSA (20 μM).

Binding-Distance Measurement Using Fluorescence Resonance Energy Transfer (FRET) between PPI and BSA

FRET is a useful spectroscopic method to delineate the structural conformations of biological and macromolecular systems such as closeness and comparative angular orientation of fluorophores, association of protein–probe composite, and so forth. The binding distance (r) between the donor (D) and the acceptor (A) can be evaluated from this useful technique.⁷² The efficiency of energy transfer (E) between D and A is manifested by the considerable overlap between the emission band of D and the absorption band of A, relative orientation of transition dipoles of D and A, and the distance between D and A, which is generally <8 nm.^73,74 In our case, Trp residue of the BSA protein acts as the donor unit, PPI acts as the acceptor unit, and the shaded portion represents the spectral overlap region between them (Figure 10). In accordance with Förster’s theory, E is defined by the following equation

19

where F₀ defines the free BSA emission intensity and F corresponds to the emission intensity of BSA in the presence of PPI. The value of R₀ (i.e., the Förster distance at which the effective transfer of energy is 50%) can be evaluated by using the following relation

20

where K² indicates the spatial orientation factor of D and A dipole. N and φ denote the refractive index of the medium and the quantum yield of D, respectively. The overlap integral between the emission spectrum of D and the absorption spectrum of A (Figure 10) is represented by J, and its value can be obtained by using the following relation

21

where, F(λ) and ε(λ) correspond to the normalized emission intensity of D in the range of λ to (λ + Δλ) and the molar extinction coefficient of A at λ, respectively.⁷⁵ In the present case, K² = 2/3, N = 1.336, and φ = 0.15.^50,76 According to eqs 19–21, parameters evaluated thereby are summarized in Table 2. The distance r between the donor unit BSA and the acceptor unit PPI after the binding interaction was <8 nm, and 0.5R₀< r < 1.5R₀ infers the occurrence of energy transfer from BSA to PPI with high possibility.⁷⁷

Figure 10.

Overlap (shaded region) between the emission spectrum of BSA and the absorption spectrum of PPI at 25 °C. [BSA] = [PPI] = 10 μM at pH 7.4. λ_ex of BSA = 295 nm.

Table 2. FRET Parameters for the BSA–PPI Composite at 25 °C.

protein	probe	J (cm3·L·mol–1)	R0 (nm)	E	r (nm)
BSA	PPI	3.17 × 10–14	3.09	0.54	3.04

Conformation Investigations: Circular Dichroism (CD) Study

The variations of the BSA secondary structure with the gradual addition of PPI were carried out by far-UV circular dichroism (CD) spectral studies in aqueous buffer solution, which show a typical profile (Figure 11) having two negative bands at ∼209 and ∼222 nm, clearly indicating the presence of an α-helix-rich secondary structure in the BSA protein.⁷⁸ These two minima in the CD spectra arise generally because of the n → π* charge transfer transition.⁷⁹Figure 11 shows a decrement in the CD signal with no significant shift of the peak position, implying the PPI-induced conformational change of the native BSA regarding the decrement of the α-helix content in BSA. The α-helix percentage in BSA can be determined by considering the following relation⁶³

22

where, the observed ellipticity values (θ_obs in mdeg at 222 nm) are used to evaluate the mean residue ellipticity (MRE) value by taking the following relation⁶³

23

where, the molar concentration of BSA is denoted by C_P. n indicates the number of amino acid residues (583 for BSA).^5−7,80,81 The cell path length is represented by l (here 1 cm). The estimated α-helix content in native BSA is found to be 65.08 (±3)%, which is in good accordance with the literature value.^5−7,80,81 A decrease in the α-helix content from ∼65.08 (±3)% in the native BSA to ∼59.93 (±3)% in the presence of 4 μM PPI (inset of Figure 11) thus obviously shows a PPI-induced perturbation of the secondary structure of BSA.

Figure 11.

CD spectral profiles of BSA (0.75 μM) with increasing concentrations of PPI at 25 °C. The inset exhibits the estimated change in the α-helix content (±3%) of BSA with the gradual addition of PPI.

Molecular Docking Results

The crystal structure analysis of BSA revealed that it is primarily composed of three homologous domains (I, II, and III), and each domain contains two subdomains (A and B).⁸² The major binding sites of BSA for various probes (exogenous and endogenous) are located in subdomains IIA and IIIA, known as Sudlow’s sites I and II, respectively.⁸³ A number of probes or drugs are available, which specifically bind either at site I or site II in BSA.^37,84,85 We have performed molecular docking study to find out whether the probe binds at site I or site II in BSA and the probable interactions involved during the association. Out of 10 different conformers, the lowest binding energy conformer was selected for analysis. The docking results are presented in Figure 12. Panel A in Figure 12 reveals that site I in subdomain IIA of BSA is the preferable binding site for the probe PPI, and this has been supported from site marker experiments. The middle panel shows the magnified view of the microenvironment around the PPI binding site in subdomain IIA of BSA near the Trp 213 residue. The possible hydrogen-bonded interaction between the nitrogen and nitrogen-bonded hydrogen atoms of PPI with Arg 217 (2.7 Å) and Asp 450 (3.2 Å) plays a crucial role in stabilizing probe-binding (Figure 12C). In addition, the probe is surrounded by various hydrophobic and polar residues. Amino acid residues such as Leu 197, Trp 213, and Val 292 provided an additional stability to the complex through hydrophobic interactions. Moreover, a number of charged and polar residues such as Arg 198, Arg 217, Lys 221, Lys 294, and so forth play a secondary role in stabilizing the PPI molecule through electrostatic interactions. Thus, docking results suggested that PPI was bound to BSA by three possible interactions, namely hydrophobic, electrostatic, and hydrogen-bonding. According to Zhang et al., increased hydrophobicity is a measure of increased stability.⁸⁶ The formation of the hydrogen bond reduces the extent of hydrophilicity and causes a significant increment in the hydrophobicity, which stabilizes the PPI–BSA complex.⁸⁷ Distance between the probe and the Trp 213 residue was 3.4 Å (Figure 12C), and the free energy (from the docking simulation) for the binding of PPI to subdomain IIA of BSA was found to be −5.01 kcal mol^–1. It is already known that Trp 134 and Trp 213 residues are responsible for the intrinsic fluorescence of BSA.³¹ The docking result illustrates that the probe binds in the near vicinity of Trp 213 in the binding pocket of site I, which in turn causes a perturbation in the fluorescence intensity of BSA, more specifically, quenches the emission of the Trp 213 residue. Thus, molecular docking study supported the experimental findings from the theoretical approach.

Figure 12.

Molecular docking of PPI with the three-dimensional structure of BSA (PDB ID: 4JK4). (A) Docking pose of PPI with BSA shown by the white circle. (B) Magnified view of the binding site of PPI in subdomain IIA. (C) Distance between the neighboring hydrogen-bonded residues and TRP 213 from the probe (PPI) molecule.

Conclusions

Here, we report a phenanthrene–pyrene-based fluorescent probe (PPI) as a molecular reporter to study the microheterogeneous environment of the BSA protein. The association between BSA and PPI has been clearly demonstrated by the UV–vis spectral change at 280 nm. The observed blue shift of the emission maximum along with an increment of the fluorescence intensity is due to the movement of PPI from a more polar aqueous environment to a more hydrophobic protein environment. The fluorescence titration of BSA with PPI resulted in a binding constant of (1.12 ± 0.06) × 10⁵ M^–1, which is in excellent agreement with the value obtained from steady-state anisotropy studies. The study on fluorescence quenching induced by PPI reveals the occurrence of both static and dynamic quenching mechanisms. The occurrence of dynamic quenching is indicated by the linear increase in ⟨τ₀⟩/⟨τ⟩ with the increasing concentration of PPI. The ground-state complexation between PPI and BSA is characterized by a large binding constant, which is very sensitive to the temperature because of a negative ΔH° value. The complexation process is also associated with a negative ΔS°, which implies that the van der Waals interactions and hydrogen-bonding interactions play the most significant roles in stabilizing the BSA–PPI complex. From the FRET study, the average distance between Trp 213 of the BSA donor and the PPI acceptor is found to be 3.04 nm, and it is close enough for nonradiative energy transfer to occur from BSA to PPI. The CD spectral studies imply PPI-induced conformational change of the native BSA in terms of decrease of the α-helix content in BSA. The site-selective binding and molecular docking studies reveal that PPI binds with BSA at site I in subdomain IIA, that is, Trp 213 is near or within the binding site of PPI. The present fluorescent probe having a planar structure could be utilized as a potential site-selective biomarker for site I in subdomain IIA.

Experimental Section

Materials

All starting materials were of reagent grade. BSA, ibuprofen, warfarin, ANS, phenanthrene-9,10-dione, and 1-pyrene carboxaldehyde were procured from Sigma-Aldrich and used as received. Deionized water from Milli-Q source was used throughout the study.

Physical Measurements

Fourier transform infrared (FTIR) spectra (4000–400 cm^–1) were recorded on a PerkinElmer RX I FTIR spectrophotometer with a solid KBr disc. The UV–vis spectral studies were recorded on an Agilent diode array spectrophotometer (Agilent 8453). Steady-state fluorescence spectra were recorded on a PTI spectrofluorimeter (Model QM-40) by using a fluorescence-free quartz cuvette of 1 cm path length. The excitation and emission slit widths were fixed at 3 nm. A Bruker 300 MHz spectrophotometer was used to run the ¹H and ¹³C NMR spectra in dimethyl sulfoxide (DMSO)-d₆ with trimethylsilane as an internal standard. The electrospray ionization mass spectra (ESI-MS⁺) (m/z) of the probe was recorded on a HRMS spectrophotometer ( QTOF Micro YA263). The time-correlated single-photon counting measurements using a picosecond diode laser (IBH Nanoled-07) in an IBH fluorocube apparatus were used to determine the fluorescence lifetimes. A Hamamatsu MCP photomultiplier (R3809) was used to collect the fluorescence decay data, which were further examined by the IBH DAS6 software. CD spectral studies were recorded on a PC-driven JASCO J815 (Japan) spectropolarimeter.

Synthesis of 2-(Pyren-1-yl)-1H-phenanthro[9,10-d]imidazole (PPI)

The probe was synthesized by the previously reported method⁸⁸ with slight modification. A mixture of phenanthrene-9,10-dione (1.04 g, 5 mmol), 1-pyrene carboxaldehyde (1.15 g, 5 mmol), and ammonium acetate (2.89 g, 37.45 mmol) were dissolved in glacial acetic acid (40 mL). Then, the resulting solution was refluxed at 110 °C for 20 h in a nitrogen atmosphere, during which time a yellowish green solid was formed. An excess of deionized water (30 mL) was added to complete the precipitation. The crude product was collected by filtration, washed with water, and dried by suction (Scheme 1). ¹H NMR (DMSO-d₆): δ in ppm 10.81 (s, 1H), 9.03 (d, 2H, J = 7.8 Hz), 8.79 (d, 3H, J = 6.63 Hz), 8.70 (d, 1H, J = 8.1 Hz), 8.60 (d, 1H, J = 7.92 Hz), 8.45 (m, 5H), 8.23 (d, 1H, J = 7.8 Hz), 7.87 (t, 4H) (Figure S1). ¹³C NMR (DMSO-d₆): δ in ppm 147.51, 133.49, 131.25, 130.68, 130.06, 129.81, 129.37, 128.98, 128.50, 127.74, 127.58, 127.15, 126.82, 125.24, 124.83, 124.28, 123.79, 123.17, 123.05 (Figure S2). ESI-MS⁺ (experimental): m/z: 419.1546 [C₃₁H₁₈N₂ + H⁺], theoretical: m/z: 419.1548 (Figure S3). IR spectrum: ν̃: 3454 cm^–1 (−NH), 1646 cm^–1 (−C=N) (Figure S4).

Experimental Solution

A 10 mM Tris buffer solution (100 mL) of pH 7.4 was prepared in deionized water, which was used in all experiments. A stock solution was prepared by dissolving the required amount of BSA (M_BSA = 66 400 g mol^–1) in pH 7.4 Tris-HCl buffer solution, and the exact concentration was determined spectrophotometrically using the molar extinction coefficient 44 000 M^–1 cm^–1 at 280 nm,⁸⁹ whereas the 10 mL stock solution of PPI (1.0 × 10^–3 M) was prepared in dimethyl formamide because of its poor solubility in water. Each solution was mixed properly before all spectral experiments at 25 °C.

Methods

All details of experimental methods are provided in the Supporting Information.

Molecular Docking Simulation Study

Molecular docking simulation studies using AutoDock (version 4.2) help to identify the probable binding site and mode of binding of the probe PPI with BSA. The RCSB Protein Data Bank (PDB ID: 4JK4) was used as a source of X-ray crystal structure of BSA. The Chem3D Ultra 8.0 was used to draw the probe structure, which was further modified using Gaussian 09W and AutoDock 4.2 programs. Gasteiger partial charges were added to the probe atoms. The nonpolar hydrogen atoms were united, and rotatable bonds were defined. Grid maps of 126 × 126 × 126 Å grid points and 0.403 Å grid spacing were generated using the AutoGrid program. The default values were used for other AutoDock parameters. The Lamarckian genetic algorithm (LGA) was used for docking calculations and the parameters were set to 100 GA runs for each docking simulation upto 250 000 energy evaluations. The population size was set to 150 with a crossover rate of 0.8 (LGA). For further analysis of docking simulations, we chose the best optimized docked model with the lowest energy, and this was best viewed in PyMOL software.

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.8b00186.

• Characterization data of compound PPI: ¹H NMR, ¹³C NMR, HRMS, and IR; Job’s plot for the binding of PPI to BSA; plot for ANS displacement study; modified Stern–Volmer plot of PPI-induced quenching of BSA; lifetime spectra of PPI in the absence and presence of BSA; lifetime spectra of BSA in the absence and presence of PPI; table for experimentally calculated quenching data and lifetime data; and experimental methods (PDF)

The authors declare no competing financial interest.