Interactions of cephalexin with bovine serum albumin: displacement reaction and molecular docking

Materials

Drug and BSA purchased from Sigma-Aldrich Co. (Poole, England) was daily used to make the stock solution of 10-⁴M. Standard stock solutions of CPL was set by dissolving a suitable amount of the pure drug (98%) in distilled water as required to 10-³M. All solutions in this study were diluted with 0.1M phosphate buffer (pH 7.4).

Apparatus

All fluorescence spectra were studied on a RF-5301 spectrofluorophotometer (Shimadzu, Kyoto Japan) and a 10 mm quartz cuvette. The widths of both the excitation and emission slits were set at 5 nm. The optimum excitation and emission wavelengths for BSA were obtained as 280 and 340 nm, respectively. To estimate and eliminate the inner filter effect (IFE), the correction ways were performed based on the absorbance amounts of solutions at excitation and emission wavelengths of albumin. UV spectra measurements were performed on a Shimadzu 2550 UV– spectrometer using a 10 mm cell at 20 nm intervals. FTIR spectra were acquired on a Shimadzu FTIR–8400 S (Shimadzu, Kyoto, Japan).

Circular Dichroism (CD) studies were made on an Aviv, model 215, spectrophotometer (Aviv Biomedical, Inc., Lakewood, NJ, USA) using 1 mm path length at room temperature with a scan rate of 500 nm/min and a reaction time of 0.5 s. The scans were repeated three times for each spectrum.

Spectroscopic investigations

All solutions must be relaxed at least 10 min before measuring the spectrum. To correct the fluorescence or absorption background, proper blanks corresponding to the buffer were deduced. The following measurable analysis was obtained by using the corrected fluorescence intensities at λ_em=340 nm at three temperatures.

Analysis of fluorescence quenching data

By using fluorescence spectroscopy, the interaction between these drugs and BSA was investigated to collect the necessary information regarding its quenching mechanism, binding constants and binding sites. Some drug molecules absorb light at the excitation and emission wavelength of BSA that influences the fluorescence intensity. It is known as the IFE. To estimate the impact of IFE in this method, the absorbance of CPL at 280 nm, excitation wavelength of BSA, and 340 nm emission wavelength of BSA were determined. The absorbance of concentrated solutions (C_CPL= 1.30×10-⁴M) was higher than 0.1, approximately about 0.25, at 2800 nm (Fig. 2). In such cases the effect of IFE should be corrected using Eq. 1.^12,13

Fig. 2. UV spectra of A: CPL (130×10-⁶M) and B: albumin (3.33×10-⁶M).

× UV spectra of A: CPL (130×10-⁶M) and B: albumin (3.33×10-⁶M).

Fig. 2.

\[{F_{cor}} = {F_{obs}} \times {e^{({A_{ex}} + {A_{em}})/2}}\] (Eq. 1)

Where F_cor and F_obs are the corrected and observed fluorescence, respectively. A_ex and A_em are the absorbance of the drug at excitation wavelength and emission wavelength, respectively.

Tryptophan residues (Trp) and tyrosine residues (Tyr) are two types of fluorophores in BSA that can be excited at 280 nm. As demonstrated and noted in Fig. 3 with the increasing amounts of CPL, the fluorescence intensities decreased without any shifts in λ_max.^14,15 If the conditions of pH, temperature, and ionic strength are fixed, two different mechanisms can occur in the fluorescence quenching, static quenching, and dynamic quenching. Stern Volmer equation can describe both of them. Normally, increasing temperature causes the increase of quenching rate constants for dynamic quenching and the reverse effect was observed for the static quenching. To configure which mechanism has an important role in the interaction, fluorescence quenching data were studied by the stern–Volmer equation (Eq. 2)¹⁶:

Fig. 3. Fluorescence emission spectra of BSA in the presence of different concentrations of CPL; [BSA] = 3.33×10-⁶M; [CPL] = (0,10,30,50,70,90,110,120 ,130)×10-⁶M (λ_ex = 280 nm, T = 293 K, pH =7.4).

× Fluorescence emission spectra of BSA in the presence of different concentrations of CPL; [BSA] = 3.33×10-⁶M; [CPL] = (0,10,30,50,70,90,110,120 ,130)×10-⁶M (λ_ex = 280 nm, T = 293 K, pH =7.4).

Fig. 3.

\[\frac{{{F_0}}}{F} = 1 + {K_{SV}}[Q] = 1 + {K_q}{\tau _0}[Q]\] (Eq. 2)

Where F₀ and F are the steady state fluorescence intensities of BSA in the absence and presence of quencher, respectively. K_SV and [Q] are the Stern–Volmer quenching constant and the concentration of the quencher, respectively. K_q is the quenching rate constant and τ₀ is the average lifetime of the BSA in the absence of any quencher and is generally equal to 10-⁸ s.¹⁷Fig. 4 shows the Stern–Volmer graphs at three different temperatures.

Fig. 4. The Stern–Volmer plots for the quenching of BSA by CPL at different temperatures, 299 K (■), 305 K (♦) and 311 K (▲) λ_ex = 280 nm, λ_em = 340 nm and pH= 7.40.

× The Stern–Volmer plots for the quenching of BSA by CPL at different temperatures, 299 K (■), 305 K (♦) and 311 K (▲) λ_ex = 280 nm, λ_em = 340 nm and pH= 7.40.

Fig. 4.

It demonstrated the plots are linear for three temperatures and the slopes decrease with increasing temperature. The values of K_SV for the interaction of CPL with BSA at three temperatures are shown in Table 1.

This study found that the K_sv values are related reversely to temperature indicating a static quenching mechanism for the interaction of BSA with CPL.¹⁸

Determination of the binding constant and the number of binding sites

There are n equivalent binding sites on the BSA and there are many methods and equations for the investigation of the binding constant, K, and the number of binding sites, n, for the reaction, P + nD→D_nP. But, only in Eq. 3, the total drug concentrations can be used¹⁹:

\[\log \frac{{{F_0} - F}}{F} = n\log k - n\log (\frac{1}{{[{D_t}] - ({F_0} - F)/[{P_t}]{F_0}}})\] (Eq. 3)

Where, F₀ and F are the fluorescence intensities in the absence and presence of the quencher, [D_t] and [P_t] are the total quencher concentration and the total protein concentration, respectively. Therefore, the number of binding sites, n, and binding constant, K, can be estimated by the plots of log [(F₀−F)/F] versus log (1/([D_t]−(F₀−F)[P_t]/F₀)), from the slopes and intercepts appropriately (Table 2). The correlation coefficients (larger than 0.98) indicates a good agreement in the interaction of CPL and BSA with Eq. 3. There are about two classes of binding site for CPL at the boundary of BSA. As demonstrated in Table 2, the K values between BSA and CPL decreased with increasing the temperature that shows the complex formation between CPL and BSA supporting the fluorescence quenching mechanism is a static process.

UV–Vis absorption spectra

The study of UV–Vis absorption spectra is a helpful method and appropriate to discover the structural change and know the formation of the complex.²⁰ For validation of the fluorescence quenching mechanism of BSA by CPL, UV–Vis absorption spectra of BSA were recorded before and after adding CPL (Fig. 5). As shown in Fig. 5, not only the absorbance intensity increased by the addition of CPL, but also the absorption spectra maximally shifted to a shorter wavelength region (279→263 nm).

Fig. 5. Absorption spectra of BSA in the presence of CPL [BSA] = 3.33×10-⁶M; [CPL] = (0,10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110)×10-⁶M, from cure a to l), the absorption spectrum of CPL only(m).

× Absorption spectra of BSA in the presence of CPL [BSA] = 3.33×10-⁶M; [CPL] = (0,10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110)×10-⁶M, from cure a to l), the absorption spectrum of CPL only(m).

Fig. 5.

FT-IR spectroscopy

FTIR spectroscopy is a simple instrument for providing structural and conformational changes of proteins. IR spectra of proteins display a number of amide bands representing diverse vibrations of the peptide moiety. Both the amide I (ranging from 1600 to 1700 cm⁻¹) and amide II (around 1548 cm⁻¹) bands of the protein are related with secondary structure of protein.

However, the change of protein secondary structure affects the amide I band more than amide II.²¹Fig. 6 shows the FT-IR spectra of free BSA and CPL–BSA in phosphate buffer. As can be noted, there is a shift in the peak position of amide I of BSA from 1652 to 1656 cm⁻¹ after addition of CPL demonstrating a slight change in the secondary structure of BSA.²²

Fig. 6. Transmittance of the buffer solution from the spectrum of the (a) BSA solution, (b) the difference spectra of BSA (subtracting the absorption of the CPL–freeform from that of CPL–BSA boundform).

× Transmittance of the buffer solution from the spectrum of the (a) BSA solution, (b) the difference spectra of BSA (subtracting the absorption of the CPL–freeform from that of CPL–BSA boundform).

Fig. 6.

Fluorescence resonance energy transfer from BSA to CPL

Fluorescence resonance energy transfer (FRET) is a powerful model that helps to evaluate donor-acceptor interactions by calculating the distance between two fluorophores, donor and acceptor.²³ The distance between two fluorophores should be shorter than 8 nm.²⁴ The energy transfer efficiency, E, is well-defined by the following Eq. (4), where r₀ is the distance from the drug to the tryptophan residue of BSA, and R₀ is the Forster critical distance, at which 50% of the excitation energy is transferred to the drug.¹² R₀ can be estimated from donor emission, BSA, and acceptor absorption spectra, drug, using the Forster equation (Eq. 5):

\[E = 1 - \frac{F}{{{F_0}}} = \frac{{R_0^6}}{{R_0^6 + r_0^6}}\] (Eq. 4)

\[R_0^6 = 8.79 \times {10^{ - 25}}{K^2}\varphi \;{N^{ - 4}}j\] (Eq. 5)

\[j = \frac{{\int_0^\alpha {F(\lambda )\,\varepsilon (\lambda ){\lambda ^4}\;d\lambda } }}{{\int_0^\alpha {F(\lambda )d\lambda } }}\] (Eq. 6)

In Eq. 5, K² is the random orientation factor associated to the geometry of the donor and acceptor and K² = 2/3 for fluid solution; N the average refractive index of medium in the wavelength range where spectral overlap is significant; Φ the fluorescence quantum yield of the donor; j the effect of the spectral overlap between the emission spectrum of the donor and the absorption spectrum of the acceptor, which could be determined by Eq. 6. Here, F(λ) is the corrected fluorescence intensity of the donor in the wavelength range λ to λ+∆λ; ε (λ) is the extinction coefficient of the acceptor at λ. FRET is an important method for studying the diversity of biological systems including energy transfer processes.

The spectral overlap between UV-Vis absorption spectrum of CPL and the fluorescence emission spectrum of free BSA is illustrated in Fig. 7. Because the fluorescence emission of BSA was affected by the excitation light around 288 nm, the spectrum ranging from 300 to 400 nm was selected to analyze the overlapping integral.

Fig. 7. The spectral overlap between UV/vis absorption spectrum of CPL (10×10-⁶M) and the fluorescence emission spectrum of free BSA (10×10-⁶M).

× The spectral overlap between UV/vis absorption spectrum of CPL (10×10-⁶M) and the fluorescence emission spectrum of free BSA (10×10-⁶M).

Fig. 7.

In this study, N= 1.36 and Φ= 0.15 according to Eqs. 4, 5 and 6. It was calculated that J = 1.97×10-¹⁵ cm³ L mol⁻¹, E = 0.0462, R₀ = 0.698 nm, and r = 1.156 nm. The average distances between BSA and CPL is less than 8 nm suggesting that energy transfer happens between BSA and CPL. Furthermore, since r is larger than R₀, it recommends that CPL quench the fluorescence of BSA by non-radiative energy transferring and static quenching.

Synchronous fluorescence spectra analysis

Synchronous fluorescence spectroscopy is a method to find evidence about the molecular environment close the chromosphere molecules. Since tryptophan (Trp) and tyrosine (Tyr) residues of BSA show the fluorescence property, we can find the specific evidence of tyrosine and tryptophan residues in BSA by setting ∆λ at 15 nm and 60 nm, respectively and determining the probable shift in wavelength emission maximum λ_max. There is no significant shift (299-301) in synchronous fluorescence spectra of BSA by addition of CPL enhancement at λ=15 nm demonstrating that CPL has slight effect on the environment of the Tyr residues in BSA. By contrast, when ∆λ=60 nm (Fig. 8), the emission maximum wavelength shows a red shift (from 342 to 335 nm) representing that the environment of the tryptophan residues were altered and hydrophobicity near this residue decreased resulting the interaction between BSA and CPL.^25-27

Fig. 8. The synchronous fluorescence spectra of BSA in the presence of CPL (T=293K, pH=7.4). [BSA]=3.33×10-⁶M; [CPL]= (0, 10, 30, 50, 70, 90, 110, 120,130) × 10-⁶M. (A) Δλ=15 nm, (B) Δλ = 60 nm.

× The synchronous fluorescence spectra of BSA in the presence of CPL (T=293K, pH=7.4). [BSA]=3.33×10-⁶M; [CPL]= (0, 10, 30, 50, 70, 90, 110, 120,130) × 10-⁶M. (A) Δλ=15 nm, (B) Δλ = 60 nm.

Fig. 8.

Circular dichroism studies

CD spectroscopy investigates the spectrum of a protein for secondary structure with highly reliability.²⁸ To study the structural change of BSA by the addition of CPL, the CD spectra of BSA was measured before and after the addition of CPL (Fig. 9). There are two negative bands in the far-UV region at ~208 and ~222 nm which are characteristic of the helical structure of proteins. As estimated, the CD spectrum displays a strong negative ellipticity at 208 nm and 222 nm.

Fig. 9. Circular dichroic spectra in the 200–240 nm range; (a) BSA, 3.33×10-⁶M; (b) BSA: CPL = 1:1; (c) 1:3; (d)1:4.

× Circular dichroic spectra in the 200–240 nm range; (a) BSA, 3.33×10-⁶M; (b) BSA: CPL = 1:1; (c) 1:3; (d)1:4.

Fig. 9.

The negative ellipticity in the region of far-UV CD was increased by the addition of the CPL to BSA (1:1, 1:3, and 1:4), without any substantial shift of peaks. This showed that the binding of CPL to BSA made an intense increase in the content of α-helical structure of the BSA clarifying the stabilization of the BSA, secondary structure as a consequence of the BSA–CPL interaction in the binary and ternary system.²⁹ This was supposed to be the result of the complex formation of BSA-CPL. However, the similarity between the shapes of the CD spectra relating to BSA before and after addition of the CPL in all relating systems proposed that the structure of BSA was still mainly α –helical.³⁰

Thermodynamic analysis for binding mode between CPL and BSA

Small particles are bound to macromolecule typically by four acting forces containing hydrogen bond, Vander Waals force, electrostatic force, and hydrophobic interaction force.

There are thermodynamic parameters dependent to temperature, enthalpy change (ΔH), entropy change (ΔS), and free energy change (ΔG) of the reaction that are significant for characterizing the binding type. Therefore, the thermodynamic parameters were investigated to describe the acting forces between CPL and BSA.

The enthalpy change (∆H°) did not differ meaningfully with the temperature range considered, and the thermodynamic factor of ∆H°, entropy change (∆S°), and free energy (∆G°) were calculated by the Van 't Hoff equation:

\[\ln K = - \frac{{\Delta {H^ \circ }}}{{RT}} + \frac{{\Delta {S^ \circ }}}{R}\] (Eq. 7)

Where K is the binding constant and R is the gas constant. The values of H° and ∆S° were evaluated using Eq. 7 by the plot of ln K versus 1/T.¹⁰ The value of ∆G° was calculated using the following equation³¹:

\[\Delta {G^ \circ } = \Delta {H^ \circ } - T\Delta {S^ \circ }\] (Eq. 8)

The calculated thermodynamic parameters for CPL- BSA interaction are illustrated in Table 2. The values of ΔG are negative indicating that the binding procedure is spontaneous. The enthalpy (ΔH) and entropy (ΔS) of the interaction of CPL and BSA are negative and positive, respectively. According to the report of Ross and Subramanian,³² the positive ΔH and ΔS value shows that the binding is mainly entropy-driven and a hydrophobic effect contributes in the interaction between CPL and BSA. The negative ΔH and ΔS values are related with hydrogen bonding and Van der Waals force. As a final point, low positive or negative ΔH and positive ΔS values are categorized by electrostatic interactions. Consequently, the interaction of CPL with BSA might contain the electrostatic interaction.

Displacement experiments using site probes and Gentamicin BSA consists of amino acid chains making a single polypeptide, which has three homologous-helices in domains (I–III). Each domain is divided into anti-parallel six helices and four sub-domains (A and B). A cluster of two sub-domains with their grooves to each other forms a domain, and three of such domains form an albumin molecule. Sites I and II are two main definite drug-binding sites in serum albumin, which are situated in particular holes in sub-domains IIA and IIIA, respectively.³³ The majority of small molecules, which are identified to combine with BSA, form a complex at site I, and only a few at site II. Though, it is challenging to find the real site from the structure of the small molecule involved. Therefore, it is proposed that site I of serum albumin demonstrated an attraction for Ketoprofen and Phenylbutazone (PB), and site II for Ibuprofen (IB) and others. To recognize the position of the CPL binding site on BSA, the displacement experiments were performed by the site probes Ibuprofen, ketoprofen, and phenylbutazon.

The percentage of fluorescence probe displaced by the drug was calculated by determining the variations in fluorescence intensity according to the method suggested by Sudlow et al.³⁴ Relative fluorescence (RF) can be used to display the alterations in fluorescence intensity in the presence of probes as Eq. 9:

\[RF = \frac{F}{{{F_0}}} \times 100\% \] (Eq. 9)

Where, F₀ and F represent the fluorescence of CPL plus BSA in the absence and presence of the probe, respectively. According to the spectral data determined in the displacement studies, the plots of F/F₀ against site probe concentration were found and are shown in Fig. 10.

Fig. 10. Effect of site marker probes (Ketoprofen ■, Phenylbutazone▲, Ibuprofen ♦) and Gentamicin (●) on the fluorescence of CPL–BSA; [BSA] = 3.33×10-⁶M; [CPL]=60×10-⁶M; [prob, Gentamicin]=(0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50) × 10-⁶M.

× Effect of site marker probes (Ketoprofen ■, Phenylbutazone▲, Ibuprofen ♦) and Gentamicin (●) on the fluorescence of CPL–BSA; [BSA] = 3.33×10-⁶M; [CPL]=60×10-⁶M; [prob, Gentamicin]=(0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50) × 10-⁶M.

Fig. 10.

The comparative fluorescence intensity meaningfully reduced after adding the phenylbutazon and ketoprofen, whereas the addition of Ibuprofen produced no observed changes, which shows that phenylbutazon and ketoprofen can displace the CPL, however, Ibuprofen has little effect on the binding of CPL to BSA. In all displacement tests, the implication was that CPL fixes to the site I in sub-domain IIA of BSA. These findings are in agreement with the results reported previously.^35,36

The reasonable procedure was examined for gentamicin, the importance of these studies is that gentamicin is classified to the aminoglycozide and are drugs typically prescribed in combination with CPL (illustrated in Fig. 10). The fluorescence demonstrated no significant change with the addition of gentamicin to the same solution, which indicates that gentamicin cannot be displaced from the binding site of CPL. Then it can be decided that gentamicin was bound to site II of BSA.

The effect of common metal ions on the binding constant

If there are metal ions present in the solution, the binding properties between CPL and BSA may be affected. A number of trace metal ions exist within the human body. Therefore, this aspect is particularly important and necessary in this research.

The effect of common ions such as Fe³⁺, Zn²⁺, K⁺ and Na⁺on the CPL– BSA binding was studied at 293 K by investigating the fluorescence intensity of CPL–BSA compound in the presence of each ion, distinctly in the range of 300–500 nm with the excitation wavelength at 280 nm. The studied cations in the phosphate buffer have no effect on the CPL–BSA interaction under the experimental conditions. The effect of properties of such cations on the interaction between a drug and BSA has been described in the former works.³⁷ The fluorescence emission spectrum of CPL in the presence of common ions (Table 3) shows no interaction between the ions and CPL. However, there is a binding reaction between the common ion and protein and thus the presence of common ion directly affects the binding between CPL and BSA.

The CPL–BSA binding constant was increased in the presence of studied ions. Therefore, the binding force between BSA and CPL was improved which extended the serum-level of CPL in the blood plasma and improved the extreme efficiency of CPL.

Molecular docking results

The lowest energy in AutoDock Vina outcome (Table 4) specified that CPL was bound to subdomain IIA better than other sites. The results of further analysis proposed that CPL interacts with BSA of the hydrophobic cavity of subdomain IIA of BSA (Fig. 11 A). CPL was surrounded by the hydrophobic and negatively charged residues such as Arg194, Arg198, Arg256, Ala290, His241, His287, and Leu237 through hydrogen bond between CPL and Arg198 (3.03A⁰, 2.92A⁰) and Arg256 (2.86A⁰) (Fig. 11 D). The existence of the hydrogen bond makes a reduction in the hydrophilicity and a rise in the hydrophobicity which becomes stable in the CPL-BSA complex.³⁸ It shows that CPL-BSA binding is mainly hydrophobic in nature. Distance between Trp213 residue and CPL was 15.9 Ǻ and free energy for the CPL-BSA complex found from the docking simulation was -6.93 kcal mol-¹ (-28.99 kj mol-¹) (Table 5), which is close to experimental ΔG of binding (-21.349 kj mol-¹) (Table 2).

Fig. 11. (A) Minimum energy docking conformation obtained from docking in subdomain (IIA) of BSA, Trp-213, Trp 134 and CPL have been identified. (B) the surface representation of BSA showing the binding pocket of CPL by PyMOL. (C) The distance between Trp-213 and CPL from docking simulation in minimum energy is shown by dotted line.(D) Representation of CPL interaction with its binding pocket in 2- dimensional view, generated by LIGPLOT.

× Minimum energy docking conformation obtained from docking in subdomain (IIA) of BSA, Trp-213, Trp 134 and CPL have been identified. (B) the surface representation of BSA showing the binding pocket of CPL by PyMOL. (C) The distance between Trp-213 and CPL from docking simulation in minimum energy is shown by dotted line.(D) Representation of CPL interaction with its binding pocket in 2- dimensional view, generated by LIGPLOT.

Fig. 11.