Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

journal homepage: www.elsevier .com/locate /saa

Specific binding and inhibition of 6-benzylaminopurine to catalase:Multiple spectroscopic methods combined with molecular docking study

1386-1425/$ - see front matter � 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.saa.2013.12.021

⇑ Corresponding authors. Tel.: +86 514 87975244.E-mail addresses: yz_xiner@163.com (Q. Xu), xyhu@yzu.edu.cn (X. Hu).

Qin Xu a,⇑, Yanni Lu a, Longyun Jing a, Lijuan Cai a, Xinfeng Zhu b, Ju Xie a, Xiaoya Hu a,⇑a College of Chemistry and Chemical Engineering, Yangzhou University, Yangzhou 225002, Chinab College of Information Technology, Yangzhou University, Yangzhou 225127, China

h i g h l i g h t s

� Fluorescence quenching mechanismbetween BLC and 6-BA was studied.� 6-BA inhibited BLC activity via a

noncompetitive manner.� Molecular docking method was used

to determine the location of 6-BAwithin BLC.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 25 June 2013Received in revised form 5 December 2013Accepted 5 December 2013Available online 25 December 2013

Keywords:Catalase6-BenzylaminopurineInteractionInhibitionDocking

a b s t r a c t

6-Benzylaminopurine (6-BA) is a kind of cytokinin which could regulate the activities of the antioxidantdefense system of plants. In this work, its interaction with and inhibition of beef liver catalase have beensystematically investigated using spectroscopic, isothermal titration calorimetric and molecular dockingmethods under physiological conditions. The fluorescence quenching of beef liver catalase (BLC) by 6-BAis due to the formation of 6-BA–BLC complex. Hydrogen bonds and van der Waals interactions playmajor roles in stabilizing the complex. The Stern–Volmer quenching constant, binding constant, thecorresponding thermodynamic parameters and binding numbers were measured. The results of UV–visabsorption, three-dimensional fluorescence, synchronous fluorescence and circular dichroism spectro-scopic results demonstrate that the binding of 6-BA results in the micro-environment change aroundtyrosine (Tyr) and tryptophan (Trp) residues of BLC. The BLC-mediated conversion of H2O2 to H2O andO2, in the presence and absence of 6-BA, was also studied. Lineweaver–Burk plot indicates a noncompet-itive type of inhibition. Molecular docking study was used to find the binding sites.

� 2013 Elsevier B.V. All rights reserved.

Introduction

The effects of cytokinin on cellular systems have received agreat deal of attention in recent decades due to the increasingapplication of cytokinin in plant culture [1]. 6-Benzylaminopurine(6-BA) (Fig. 1), as one of the first-generation synthetic cytokinins,was used to stimulate cell division, lateral bud emergence (apples,

oranges), basal shoot formation (roses, orchids), flowering (cycla-men, cacti) and fruit set (grapes, oranges, melons) [2,3]. It is alsoregarded as a good candidate for postharvest applications and usedas a biopesticide in the USA and Canada [4]. Recent research showsthat 6-BA would delay the production of reactive oxygen species,and regulate the oxidative status of tissues. It has a close relation-ship with the antioxidant defense system of plants [5–7].

Catalase is one of the most important proteins of the antioxi-dant defense system whose function is to protect cells from thetoxic effects by catalyzing the decomposition of hydrogen peroxide

Fig. 1. Structure of 6-benzylaminopurine.

328 Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335

into molecular oxygen and water. Recently, there is increasingevidence that catalase is a major factor in a variety of pathologicalstates such as diabetes, aging, oxidative stress, and cancer [8]. In-take of any extraneous chemical is likely to enhance or inhibitthe catalytic activity of catalase in vivo. Understanding the interac-tion mechanism between catalase and various extraneous chemi-cals and how these interactions influence their biologicalfunctions would help us guide the proper use of these chemicals.In recent years, some researches on the interactions between cat-alase and chemicals have been developed. Sarmiento et al. havestudied the interaction between nafcillin and catalase by equilib-rium dialysis and f-potential measurements [9]. But this procedurecould not provide the conformation changes information. Fluores-cence spectroscopy can provide particularly useful information onbinding modes by a simple procedure [10]. Liu et al. have reportedthe interaction mechanisms between catalase and different chem-icals such as 4-aminoantipyrine [11], oxytetracycline [12] and flu-oroquinolones [13] and characterized their harmful effects byspectroscopic and computational methods. 6-BA has been reportedto adjust catalase activities. Díaz–Vivancos found that catalaseactivity in crocus sativus explants was reduced in the presence6-BA [14]. Zavaleta–Mancera et al. reported that 6-BA caused theincrease of the activities of catalase in the delayed-senescence tis-sue in dark [15]. However, all of the reported works studied theregulation of catalase activity by 6-BA in plants or plant tissues.Up to now there was no evidence whatsoever supporting the ideathat 6-BA would interacts with catalase. Studying the interactionmechanisms between 6-BA and catalase and the structure changesof catalase in the presence of 6-BA could help us to better under-stand the influence of 6-BA in vivo at the molecular level and guidethe proper use of 6-BA.

In this work, the interaction between 6-BA and a kind of cata-lase, beef liver catalase (BLC), has been investigated by UV–visspectroscopy, fluorescence spectroscopy, circular dichroism (CD)spectroscopy, isothermal titration calorimetry (ITC) and moleculardocking methods. The thermodynamic functions, energy transferefficiency between the donor and acceptor, binding distances,and effect of 6-BA on the conformation of BLC were investigated.The inhibitory effects and inhibition type of 6-BA on BLC wereevaluated. We also tried to locate the best binding sites and to con-struct the binding modes according to the molecular dockingresults.

Experimental

Reagents and solutions

Crystalline BLC (molecular weight 240 kD, Product No. C-9322,activity P2000 U/mg, from Bovine Liver, Sigma) was directlydissolved in water to prepare stock solutions (2.0 � 10�5 M) andstored at 4 �C. 6-BA was purchased from Sangon Biotech in

Shanghai (China). The stock solution of 6-BA was prepared by dis-solving it in ethanol to a suitable concentration (6.0 � 10�3 M).50 mM Tris–HCl buffer containing 0.1 M NaCl (pH 7.4) was usedfor all experiments. All chemicals were used without furtherpurifications.

UV–vis absorption and fluorescence experiments

All UV–vis absorption spectra were measured on a ShimadzuUV-2550 spectrophotometer (Shimadzu, Kyoto, Japan) by using1 cm quartz cells.

All fluorescence spectra were recorded on an F-4500 fluores-cence spectrophotometer (Hitachi Japan) equipped with a 1 cm cellquartz cuvette, a thermostat bath (Model ZC-10) and a Xenonlamp. The excitation and emission slit widths were both set at5 nm. The scanning speed was 1200 nm/min.

The fluorescence intensities used in this work were corrected toaccount for the inner filter effect by using Eq. (1) [16]:

Fcorr ¼ Fobs � 10AexdexþAemdem

2 ð1Þ

where Fcorr and Fobs are the corrected and observed fluorescenceintensities respectively. Aex and Aem are the absorbance values atexcitation and emission wavelengths respectively, and dex and dem

are the cuvette path lengths in the excitation and emission direction(in cm) respectively.

In a typical fluorescence measurement, 1.0 mL of 1.0 � 10�5 MBLC solution, and different concentrations of 6-BA were added intoa 10 mL colorimetric tube, successively. The samples were dilutedto scaled volume with 0.1 M pH 7.40 Tris–HCl, mixed thoroughlyby shaking, and kept static for 15 min at different temperatures(293, 303, and 313 K). The excitation wavelength for BLC was280 nm, with the excitation and emission slit widths set at 5 nm.The solution was subsequently scanned on the fluorophotometerand the fluorescent intensity at 350 nm was determined. The syn-chronous fluorescence spectra were recorded when the Dk valuebetween the excitation and emission wavelengths was stabilizedat 15 and 60 nm, respectively.

To confirm the quenching mechanism, the fluorescence quench-ing data were analyzed using the Stern–Volmer equation (Eq. (2))[16].

F0

F¼ 1þ KSV½Q � ¼ 1þ kqs0½Q � ð2Þ

where, F0 and F are emission intensities of BLC in the absence andpresence of 6-BA respectively, KSV is the Stern–Volmer constant re-lated to the bimolecular quenching rate constant (kq) by KSV = kq�s0,and s0 is the excited state lifetime of BLC which is 10�8 s [17], and[Q] is the concentration of 6-BA.

For the binding-related quenching process, the binding constantKa and Hill’s coefficient n can be determined on the basis of the fol-lowing Eq. (3) [18]:

lg½ðF0 � FÞ=F� ¼ lgKa þ nlg½Q � ð3Þ

This form of Hill’s equation is used to estimate the cooperativityin multi-subunit BLC.

The thermodynamic parameters, enthalpy changes (DHh), andentropy changes (DSh), can be calculated from the thermodynamicVan’t Hoff equation:

lnKa ¼�DHh

RTþ DSh

Rð4Þ

where Ka is the associate constant at the corresponding tempera-ture and R is gas constant. The temperatures used were 298, 308,and 318 K. There is a linear relationship between lnKh and 1/T.The enthalpy change (DHh) can be obtained by the slope, and DSh

Fig. 2. UV–vis spectra of beef liver catalase (BLC) in the absence and presence ofdifferent concentrations of 6-BA in pH 7.4 Tris–HCl buffer (vs. the same concen-tration of 6-BA solution) (a)1.0 lM BLC; (b)10.0 lM 6-BA + 1.0 lM BLC; and (c)50.0 lM 6-BA + 1.0 lM BLC.

Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335 329

from the intercept. The free energy change (DGh) can be obtainedfrom the following relationship:

DGh ¼ DHh � TDSh ð5Þ

Föster’s energy transfer theory [19] has been used to calculatethe distance between the protein residues (donor) and 6-BA(acceptor) in the binding site. By Förster’s theory, the efficiencyof energy transfer (E) related to the distance R0 between donorand acceptor following Eq. (5):

E ¼ 1� FF0¼ R6

0

R60 þ r6

ð6Þ

where r is the distance between the donor and the acceptor and R0

represents the critical distance when the transfer efficiency equals50%. R0 can be calculated from Eq. (7):

R60 ¼ 8:8� 10�25K2N�4UJ ð7Þ

where K2 is the orientation factor, N is the refractive index of themedium, U is the quantum yield of the donor. The spectral overlapintegral (J) between the donor’s emission and the acceptor’s absorp-tion spectrum is calculated by the following Eq. (8):

J ¼ RFðkÞeðkÞk4dkRFðkÞdðkÞ ð8Þ

where F(k) is the fluorescence intensity of the donor when thewavelength is k and e(k) is the molar absorbance coefficient of theacceptor at the wavelength of k. The value of J can be calculatedby integrating the overlap of the UV absorption spectrum of 6-BAwith the fluorescence emission spectrum of BLC.

Circular dichroism measurements

The circular dichroism spectra of BLC (1.0 � 10�6 M) were col-lected in the presence of 6-BA at molar ratio of 1:0 and 1:10 in aJASCO (J-810) spectropolarimeter. Spectra were collected from200 to 250 nm with 200 nm/min scan speed and a response timeof 1 s. Respective blanks were subtracted. Each CD spectrumobtained was the average of three scans.

Isothermal titration calorimetry (ITC) measurements

ITC was performed with a nanoITC SVH calorimeter (TAinstruments, USA) at 25 �C. Samples were buffered at pH 7.4 with50 mM Tris–HCl solution. In a typical experiment, BLC solution(5.0 � 10�6 M) was placed in the sample cell of the calorimeterand 6-BA solution (8.0 � 10�4 M) was loaded into the injectionsyringe. A typical titration was performed by sequential injectionsof 6 lL of 6-BA solution into BLC sample (950 lL). The sample cellwas stirred continuously at 350 rpm. The time delay betweensuccessive injections was 300 s. Raw data were obtained as a plotof heat (lJ) against injection number. Control experiments in-cluded the titration of 8.0 � 10�4 M 6-BA into buffer, buffer intoBLC and buffer into buffer. The last two controls resulted in smalland equal enthalpy changes for each successive injection of buffer,and therefore, were not further considered in the data analysis.Corrected data refer to experimental data after subtraction ofthe 6-BA into buffer control data. The first injection was ignoredin the final data analysis. Integration of the peaks correspondingto the injections and correction for the baseline were carriedout using Nano-Analyze program (TA-Instruments). All mea-surements were repeated three times. Since TA Instrumentsuse a positive sign for an exothermic reaction, which is oppositeto the conventional presentation used by MicroCal (GE), we

manually converted the measured raw data that are presented inall figure.

BLC activity assay and inhibition study

BLC activity was determined by detecting H2O2 consumptionaccording to the procedure by Beers and Sizer [20]. In a total vol-ume of 3 mL sample, 1.0 � 10�7 M of BLC was incubated togetherwith 0.018 M H2O2 and different concentrations of 6-BA in50 mM Tris–HCl buffer (pH 7.4). The absorbance of the reactionmixture at 240 nm was recorded for 1 min every 10 s. One unitof BLC activity is defined as the amount of enzyme that decom-poses 1.0 mM H2O2 per min. Controls incubated without 6-BA werealso included and the results were expressed as the relativepercentage of activity in respect to the control.

For inhibition studies, a matrix of substrate (H2O2) between0.005 and 0.03 M and different concentrations of inhibitor (6-BA)was generated. Values of the absorbance of H2O2ðAH2OÞ weredetermined after short reaction times in triplicate at each H2O2

concentration. The Lineweaver–Burk plot of 1=AH2O2 against1/[H2O2] was used to identify the inhibition type [21].

Molecular docking

Molecular docking was performed to obtain protein–ligandbinding energy and to identify potential ligand binding sites. Dock-ing calculations were carried out with AutoDock 4.2 and AutoDockTools (ADT) software using the Lamarkian Genetic Algorithm (LGA)based on the adaptive local search method [22]. The crystalstructure of BLC (PDB code 1TGU) was downloaded from theProtein Data Bank [23]. The 3D structure of 6-BA was generatedby GaussView 5.08. The minimized energy conformation of 6-BAwas obtained by Gaussian 09 software. For reorganization of thebinding sites in BLC, blind docking was carried out with settingof grid box size 126 Å � 126 Å � 126 Å along x, y, z axes with a gridspacing of 0.375 Å after assigning the protein and probe withKollman charges. Local search was used to identify the optimumbinding site of 6-BA in BLC. AutoGrid was run to generate a gridmap of various atoms of the ligand and receptor. After the comple-tion of grid map, ligand flexible docking simulations wereperformed with 100 GA population size. Finally, the conformationwith the lowest binding free energy was used for further analysis.PyMOL was used for visualizing the interaction of dockedprotein–ligand complex.

Table 1Stern–Volmer quenching constants for the interaction of 6-BA with beef liver catalaseat different temperatures in pH 7.4 Tris–HCl buffer.

pH T (K) Ka (�104 L mol�1) Kq (�1012 L mol�1 s�1) Ra S.Db

7.4 298 0.7681 0.7681 0.9984 0.009308 0.7105 0.7105 0.9876 0.015318 0.6430 0.6430 0.9940 0.010

a R is the correlation coefficient.b S.D. is the standard deviation for the KSV values.

330 Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335

Result and discussion

Effect of 6-BA on the absorption spectra of BLC

UV–vis absorption measurement is a very simple method appli-cable to explore structural changes of BLC in the presence of 6-BA.Fig. 2 shows the UV–vis absorption spectra of BLC in the absence(curve a) and presence of different concentrations of 6-BA (curveb and c). Two absorption peaks are observed in the studied wave-length range. The peak at about 270 nm is from the aromatic aminoacids of BLC, i.e., tyrosine (Tyr), tryptophan (Trp) and phenylala-nine (Phe). The changes observed in this region give evidence ofthe rearrangement in the globule structure of BLC, as reported be-fore for BLC in different pH solutions [24]. The peak at about407 nm (Soret band) arises from the p–p� transitions of the hemesystem of BLC. This Soret band is very sensitive to variation of themicroenvironments around the prosthetic groups. With the addi-tion of 6-BA into BLC solution, an increase of the absorbance at270 nm along with a slight blue-shift of kmax is observed (Fig. 2).The shift, according to the literatures [24,25], indicates the exten-sion of the peptide strands of the protein and the formation of 6-BA–BLC complex. The subtle change of the peak around 407 nmsuggests that the microenvironment around the heme changes alittle.

Effect of 6-BA on the fluorescence spectra of BLC

Fluorescence quenching is a powerful tool to characterize theaccessibility of fluorophores in the protein matrix to quenchersin solution. Detailed information such as binding sites, dynamicsand conformational transformation can be obtained by using fluo-rometric techniques. The intrinsic fluorescence of BLC mainlycomes from Trp and Tyr residues and is very sensitive to theirmicroenvironment. Therefore, we took advantage of fluorescencespectroscopy to investigate the interaction between 6-BA andBLC. Fig. 3 shows the fluorescence spectra of BLC in the absenceand presence different concentrations of 6-BA. The fluorescenceintensity of BLC decreases continuously with the successive addi-tion of 6-BA, indicating that the binding of 6-BA to BLC alters themicroenvironment around the fluorophores and quench the intrin-sic fluorescence of BLC.

Fluorescence quenching can be caused by collisional or binding-related quenching process. Since higher temperatures results infaster diffusion and hence leads to larger amounts of collision

Fig. 3. Effect of 6-BA on BLC fluorescence (kex = 280 nm). Inset is the Stern–Volmerplots for the quenching of BLC by 6-BA at 298 K. BLC: concentration: 1.0 lM; 6-BA/(�10�5 M): (a) 0.0; (b) 1.0; (c) 2.0; (d) 3.0; (e) 4.0; and (f) 5.0, pH 7.4, T = 298 K.

quenching, the constants for collision quenching of the fluorescentcomplexes will increase with temperature. In contrast, increasedtemperature is likely to result in decreased stability of complexes,and thus the binding-related quenching constants are expected todecrease with increasing temperature. In order to interpret thequenching mechanism between 6-BA and BLC, the fluorescencequenching constant (KSV) is measured according to the well-knownStern–Volmer equation (shown in Experimental section, Eq. (2)) atdifferent temperatures. Inset of Fig. 3 is the Stern–Volmer plot ob-tained at 298 K. The good fitting linearity with R-square value>0.99 not only suggests that the Stern–Volmer model is appropri-ate for studying the binding mechanism between 6-BA and BLC,but also implies that only one dominant quenching pathway existsduring the titration process [16]. The calculated quenching con-stants of KSV and Kq at different temperatures were listed in Table 1.The KSV values are decreased with the increase of temperatures,indicating the weakening of the 6-BA–BLC complex. This resultconfirms that the fluorescence quenching of BLC induced by 6-BAmay be initiated by binding-related complex formation rather thanby dynamic collision. Furthermore, the values of Kq listed in Table 1.are larger than the maximum scatter collision quenching constant(2 � 1010 M�1 s�1) [26]. This also suggested that the principalmechanism for the fluorescence quenching was due to the forma-tion of 6-BA–BLC complex.

Bind constants, binding numbers, thermodynamic parameters andbinding modes of 6-BA–BLC systems

As discussed above, the fluorescence of BLC is quenched by 6-BAbinding via the complex-formation dominant process. A greatnumber of different equations can be used to analyze the quench-ing data, but the most used one to determine both the bindingconstant and the number of binding sites n is a double log Stern–Volmer equation. As was pointed out [18], the double log plottinglinearized non-linear data, resulting in a false impression. How-ever, the double log Stern–Volmer equation (Eq. (3)) can be usedto fit the data even if the hypothesis of infinite cooperativity isnot satisfied [18]. In this case, ‘‘n’’ is the Hill coefficient, which islower than the real number of binding sites. It is a measure ofthe cooperativity between the binding sites, not the number ofbinding sites. Table 2. gives the calculated Ka and n analyzed byEq. (3). For each temperature, the binding constant (Ka) of 6-BA–BLC is at 103 order and the number of Hill coefficient (n) approxi-mates to 1. The decrease of Ka values with increased temperaturesfurther suggested that the quenching was a complex-formationprocess; hence it led to the reduction of the stability of 6-BA–BLCsystems. The value ‘‘n’’ close to 1 suggests that the binding sitesof BLC are noncooperative.

Isothermal titration calorimetry (ITC) is a technique used tomeasure the heat exchange associated with molecular interactionsat constant temperature. It could provide binding stoichiometry(n). A representative calorimetric titration profile of BLC with 6-BA at pH 7.4 and 298 K is shown in panel A of Fig. 4. Each peakin the isotherm represents a single injection of 6-BA into BLC solu-tion. The negative heat deflection indicates that the binding is an

Table 2Binding constants Ka and relative thermodynamic parameters of the 6-BA–BLC system.

T (K) Ka (�104 L mol�1) n Ra DGo (kJ mol�1) DSo (J mol�1 K�1) DHo (kJ mol�1)

298 0.7155 0.99 0.9960 �22.17308 0.6296 0.99 0.9957 �22.68 �51.05 �6.96318 0.5305 0.98 0.9930 �23.19

a R is the correlation coefficient.

Fig. 4. Isothermal titration calorimetry for the interaction between 6-BA and BLC at298 K, pH 7.4. Panel A shows heat flow for each injection (lcal/s) as a function oftime (minutes); Panel B shows integrated heats in each injection by subtracting theheat of dilution of 6-BA. The solid line represents the best fit obtained from multiplesites model.

Fig. 5. Synchronous fluorescence spectra of BLC at different concentrations of 6-BAwhen (A) Dk = 15 nm; and (B) Dk = 60 nm. The concentration of BLC is 1.0 lM. Theconcentrations of 6-BA (10�5 M) are: (a) 0; (b) 1.0; (c) 2.0; (d) 3.0; (e) 4.0; and (f)5.0. The arrow indicates the direction of increase of the 6-BA concentration. Thespectra are recorded at pH 7.4, 298 K.

Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335 331

exothermic process. The panel B of Fig. 4 shows the plot of theamount of heat liberated per injection as a function of the molarratio of 6-BA to BLC. The stoichiometry (n) can be analyzed byfitting the data to various interaction models in the NanoAnalyzesoftware of TA Instruments. When the multiple sites model wasused to study the interaction between 6-BA and BLC, the coinci-dence degree between calculated curve and experimental integrateheat was most reasonable (panel B, Fig. 4). The data reveals thateight 6-BA binds to one BLC.

Generally, the binding forces between ligands and biomoleculesmay comprise hydrogen bonds, electrostatic forces, van der Waalsforces and hydrophobic interactions, etc. The thermodynamicparameters, free energy changes (DGh), enthalpy changes (DHh),and entropy changes (DSh), are the main evidences for confirmingbinding modes. They could be calculated according to the van’tHoff equation (Eqs. (4) and (5)). From the viewpoint of thermody-namics, DHh > 0 and DSh > 0 imply that hydrophobic interaction isthe main force; DHh < 0 and DSh < 0 reflect van der Waals force orhydrogen bonding; DHh < 0 and DSh > 0 suggest electrostatic forcesplay a key role [27]. On account of the little temperature effect,DHh can be regarded as a constant if the temperature range isnot too wide. The calculated thermodynamic parameters and Ka

values between 6-BA and BLC using fluorescence spectroscopicmethod are listed in Table 2. The negative value of free energy(DGh) supports the assumption that the binding process is sponta-neous. The corresponding negative entropy DHh and enthalpy DSh

values suggest that BLC and 6-BA form a complex, and decrease thenumber of molecules of the system. This process is the complexformation quenching. Hydrogen bonds and van der Waals interac-tions play major roles in the reaction.

Conformation changes of BLC on interaction with 6-BA

Synchronous fluorescence spectrum (SFS) is obtained by a syn-chronous motion of an excitation and emission monochromator atthe same rate, but with a constant difference of wavelength Dk. It

has been widely used to characterize fluorescence properties ofprotein–ligand system, and to provide further information aboutthe microenvironment changes of biomolecules based on measure-ments of the possible shift in maximum emission wavelength [28].When Dk between the excitation wavelength and the emissionwavelength is 15 nm, the SFS offers characteristics correspondingto the polarity changes around tyrosine (Tyr) residues. When Dkis 60 nm, the SFS is characteristic of tryptophan (Trp) residues[29]. The SFS of BLC upon addition of various amounts of 6-BA isdisplayed in Fig. 5. The excitation wavelengths of Tyr and Trp res-idues are red-shifted with increasing concentration of 6-BA, andthe fluorescence intensities of BLC decrease. The red-shift impliesthat the binding of 6-BA to BLC is close to Tyr and Trp residues.The micro-environmental surroundings of Tyr and Trp residueshave slight changes in the presence of 6-BA. The fluorescencequenching of BLC by 6-BA is the result of the formation of6-BA–BLC complex, as reported before for a bovine serumalbumin-6-mercaptopurine complex [30].

Three-dimensional fluorescence contour spectrum is an analy-sis technique that will give additional information or evidenceregarding conformational changes of BLC in the presence of 6-BA.Fig. 6 presents the three-dimensional fluorescence contour spectraof BLC (A), 6-BA–BLC (B) and 6-BA (C), respectively. Peaks 1 and 2are the Rayleigh scattering peak (kex = kem) and the second-orderscattering peak (kem = 2kex), respectively. In Fig. 6A, peak A at kex/kem = 280/349 nm reveals the spectral behavior of Trp and Tyr res-idues; and is closely related to the microenvironmental polarityaround these residues. Peak B at kex/kem = 234/346 nm mainly re-flects the fluorescence character of the polypeptide backbonestructure. Fluorescence peak of 6-BA is at kex/kem = 231/341 nmand its intensity is very weak (Fig. 6C). In the absence of 6-BA,the relative fluorescence intensity of peaks A and B are 1162 and255, respectively (Fig. 6A). After the addition of 6-BA, the relativefluorescence intensity of peaks A and B decrease to 771 and 219,respectively, and peak A shifted to kex/kem = 289/349 nm (Fig. 6B).The small decrease of peak B intensity with the addition of 6-BA

Fig. 6. Three-dimensional fluorescence contour spectra of BLC (A), 6-BA–BLC (B), and 6-BA (C). The concentration of BLC is 1.0 lM. The concentrations of 6-BA is 5.0 lM. Thespectra are collected at pH 7.4, 298 K.

Fig. 7. CD spectra of the 6-BA–BLC system. (a) 1.0 lM BLC, and (b) 1.0 lMBLC + 20.0 lM 6-BA.

Fig. 8. Overlapping between the fluorescence emission spectrum of BLC (a) and UVabsorbance spectrum of 6-BA and (b). The concentration of BLC is 1.0 lM and 6-BA10.0 lM. The spectra are collected at pH 7.4, 298 K.

332 Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335

indicates that the peptide structure of BLC has been changedslightly. The fluorescence intensity of peak A has been quenched33.6%, suggesting that 6-BA binds to BLC, and the binding site isnear the Trp and Tyr residues. Some micro-environmental and con-formational changes in BLC happened, and a complex between 6-BA and BLC formed, which is identical with the previous studieson the interaction between cinnamic acid and lysozyme [31].

To get deep insight into the possible influence of 6-BA bindingon the secondary structure of BLC, far-UV CD measurements ofBLC were performed in the absence and presence of 6-BA, as shownin Fig. 7. The CD analyses for four secondary structures, a-helix, b-sheet, b-turn and random coil, were performed by using the CDPro

software package. BLC has the secondary structures of 26.0% a-he-lix, 24.2% b-sheet, 25.4% b-turn and 24.4% random coil. With theaddition of 6-BA to BLC, the a-helix increased to 29.7%, the b-sheetdecreased to 22.6%, and the b- turn increased to 26.2%. The resultsindicated that 6-BA bound to the amino acid residues of BLC, andfurther caused some secondary structure changes of BLC.

Energy transfer between 6-BA and BLC

Föster’s energy transfer theory [19] has been used to calculatethe distance between the protein residues (donor) and 6-BA

Fig. 9. (A) Changes of BLC activity with the increase of the concentration of 6-BA. Conditions: 3 mL assay solution containing 50 mM Tris–HCl buffer (pH 7.4), 18 mM of H2O2,0.1 lM of BLC and different concentrations of 6-BA at 298 K. (B) Double-reciprocal Lineweaver–Burk plot. Results are mean ± S.D. of three independent experiments. (a)0.1 lM BLC; (b) 0.1 lM BLC + 4.0 lM 6-BA; and (c) 0.1 lM BLC + 8.0 lM 6-BA. The data are collected at pH 7.4, 298 K. (C) Schematic diagram of the interaction between 6-BAand BLC.

Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335 333

(acceptor) in the binding site according to Eqs. (6)–(8). The value ofJ can be calculated by integrating the overlap of the UV absorptionspectrum of 6-BA with the fluorescence emission spectrum of BLC(Fig. 8). Using K2 = 2/3, n = 1.36, U = 0.15 [32], we were able to cal-culate that J = 2.262 � 10�15 cm3 mol�1, R0 = 2.02 nm, E = 37.2%and r = 2.21 nm according to Eqs. (5)–(7). The values of R0 and rare lower than 7 nm. This represents an indication that indicatingthat energy transfer is likely to occur between 6-BA and BLC, as re-ported for guaiacol-human serum albumin complex [33].

Effect of 6-BA on BLC activity

Catalase plays a major role in protecting tissues from toxic ef-fects of H2O2 and partially reduced oxygen species. Its activity isinfluenced by some compounds [34]. The effect of 6-BA on theactivity of BLC was also studied in this work. Fig. 9A shows thealterations in enzymatic activity of BLC as a function of 6-BA con-centration. The activity of BLC decreases with increasing amount of6-BA in vitro under physical conditions, and is stabilized to about0.9% of its original value when the concentration of 6-BA is morethan 20.0 lM. The result suggests that the activity of BLC decreasedin the presence of 6-BA, which may be caused by the conforma-tional changes [35] as reported before for chrysoidine and catalasecomplex [36]. The finding that 6-BA inhibits catalase may appear incontradiction with most literature on the subject stating that 6-BAactually increases the activity of catalase and some other antioxi-dant enzymes in vivo or ex vivo studies [37,38]. Some work

suggested that loss of catalase activity might amplify oxidativestress [39]. It is plausible that the partial inhibition of catalase by6-BA triggers a strong anti-oxidative response in live cells resultingin larger catalase expression. Thus it seems that 6-BA increase theactivity of catalase.

The double-reciprocal (also known as the Lineweaver–Burk)plot was used to study the inhibition type. Fig. 9B shows a seriesof lines converging on the same point on the x while the y-inter-cept of the plots increases with the increase of 6-BA concentration.For Lineweaver–Burk plot, the slope of the resulting line is KM/Vmax,the y-intercept is 1/Vmax, and the x-intercept is �1/KM. The inhibi-tion of BLC by 6-BA causes a decrease in Vmax value while KM isunaffected. The hallmarks of noncompetitive inhibition are anunchanging Michaelis constant (KM) and a decrease of the maxi-mum velocity (Vmax) when the inhibitor is present [40]. So 6-BAacts as a noncompetitive inhibitor by binding to BLC. Fig. 9C wasused to illustrate the inhibition mechanism. 6-BA binds to BLC ata location other than the active site. The Km calculated from thex-intercept of Fig. 9B is 70.02 mM, which is close to the reportedvalue of 69.96 mM [41].

Molecular docking

Solution experiments may be more closely represent physiolog-ical conditions, but it is difficult to locate the binding site. Molecu-lar docking software was further employed to find the binding sitesof 6-BA on BLC to better understand their interaction. Among the

Table 3The lowest energy-ranked results of five 6-BA–BLC binding conformations.

Energy-ranked results Conformation data

1 2 3 4 5

Binding energy (kcal mol�1) �7.28 �7.21 �7.14 �7.07 �7.04Ligand efficiency (kcal mol�1/non-hydrogen atom) �0.43 �0.42 �0.42 �0.42 �0.41Inhibition constant (lM) 4.58 5.15 5.89 6.59 6.92Intermolecular energy (kcal mol�1) �8.18 �8.11 �8.03 �7.96 �7.85

Fig. 10. (A) Docking conformations of 6-BA in complex with BLC. (B) Bindingmodels of 6-BA. The short red dotted line refereed to the hydrogen bonds. The longred dotted line refereed to the distance between 6-BA and Trp 142 (Å). (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of this article.)

334 Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335

various conformers of docking results, 5 were chosen on the basisof the free binding energy and score ranking. Data related to themare reported in Table 3. [42]. The minimum energy conformationstate of 6-BA bound protein complex in our study was consideredas the best one, and no effort was done to locate the second site.The minimum binding energy conformer is shown in Fig. 10. Frido-vich et al. have indicated that heme is the active-site of catalase[43]. Fig. 10A shows that the most-favorable binding site of 6-BAis located away from heme groups of BLC, compatible with the ideathat 6-BA is a noncompetitive inhibitor. Fig. 10B shows that theprobe molecule is surrounded by Gly 39, Asp 58, His 62, Arg 362,Pro 367, Asn 368, Tyr 369, Leu 370, Gln 371, Asn 384, Gln 386,Arg 387, Asp 388, Cys 392, Asn 396 and Gln 397. This can be usedto explain the red-shift of the excitation wavelength of Tyr residueswhen Dk is set at 15 nm. Five hydrogen bonds between N2, N4 andN6 atoms of 6-BA and the adjacent H atom of the hydroxyl groupsof Asn 368, Asp 386, Asn 396 and Gln 386 are observed. The forma-tion of hydrogen bonds stabilizes the complex of 6-BA and BLC andcontributes to the 3D space position change of 6-BA to adapt thebinding site of BLC. Docking results also show the presence ofvan der Waals interactions. The hydrogen bonds and van der Waalsare the most important factors contributing to the observed shift ofthe wavelength and hence to the stability of protein associationcomplexes [44].

From the docking simulation, the observed free energy changeof binding (DG) for the complex 6-BA–BLC is found to be�30.45 kJ mol�1, which is close to our experimental bindingenergy (�22.17 kJ mol�1) obtained from complexation study bythermodynamic Van’t Hoff equation. The distance between Trp142 and the bound probe obtained from docking simulation(1.4 nm) is in the FRET length range, which might cause the red-shift of the excitation wavenumber of Trp residues when Dk is60 nm (Fig. 5B).

All results obtained via the docking are identical to the experi-ment results. It is reasonably concluded that the 3D structure of

BLC designed is reasonable and suitable for docking study, anddeep insight into the binding mode of 6-BA and BLC has been suc-cessively explored.

Conclusion

The present work reports the interaction of 6-BA to BLC. Resultsindicate that 6-BA can spontaneously bind with BLC mainlythrough hydrogen bonds and van der Waals interactions. The mi-cro-environment change around tyrosine (Tyr) and tryptophan(Trp) residues of BLC occurs in the presence of 6-BA as revealedby UV–vis absorption and CD spectroscopy studies. ITC resultsshow that eight 6-BA binds to one BLC. The best binding sites of6-BA within BLC was determined by molecular-docking experi-ments. The interaction between 6-BA and BLC induces an obviousloss in catalytic activity of BLC and 6-BA inhibits the activity ofBLC via a noncompetitive manner. This work provides an accurateand full basic data for clarifying the binding effects of 6-BA on BLC,and also contributes to understanding the mechanism of oxidativestress caused by the bound 6-BA from the functional macromolec-ular level. The methods used in this work could also be applied toexplore the interaction mechanism between other cytokinins andcatalase.

Acknowledgments

The authors gratefully acknowledge the financial support by theNational Natural Science Foundation of China (21075107,21275124 and 21275125), the Foundation of Jiangsu EducationalBureau (12KJB150022) and the Qinglan Project of JiangSu Province(Grant No. 11KJB150019), and the project funded by the PriorityAcademic Program Development of Jiangsu Higher Education Insti-tutions and Natural Science Foundation of Jiangsu Province.

References

[1] S. Perilli, L. Moubayidin, S. Sabatini, Curr. Opin. Plant Biol. 13 (2010) 21–26.[2] M.C. Polanco, M.I. Pelaez, M.L. Ruiz, Plant Cell Tissue Organ. Cult. 15 (1988)

175–182.[3] G.F. Barclay, C.R. McDavid, Sci. Hortic. 72 (1998) 81–86.[4] R. Prange, A. Ramin, B. Daniels-Lake, J. DeLong, P. Braun, HortScience 41 (2006)

301–303.[5] B.X. Chen, H.Q. Yang, J. Sci. Food Agric. 93 (2013) 1915–1921.[6] F. Xu, Z. Yang, X. Chen, P. Jin, X. Wang, Y. Zheng, J. Agric. Food. Chem. 60 (2012)

234–240.[7] M. Costa, P. Civello, A. Chaves, G. Martínez, Postharvest Biol. Technol. 35 (2005)

191–199.[8] A. Gonenc, A. Hacisevki, S. Aslan, M. Torun, B. Simsek, Eur. J. Int. Med. 23 (2012)

350–354.[9] J.M. Ruso, P. Taboada, P. Martínez-Landeira, G. Prieto, F.l. Sarmiento, J. Phys.

Chem. B 105 (2001) 2644–2648.[10] M.R. Eftink, Methods Enzymol. 278 (1997) 221–257.[11] Y. Teng, H. Zhang, R. Liu, Mol. Biosyst. 7 (2011) 3157–3163.[12] Z.X. Chi, R.T. Liu, H. Zhang, Sci. Total Environ. 408 (2010) 5399–5404.[13] Q.P. Qin, R.T. Liu, J. Hazard. Mater. 252–253 (2013) 321–329.[14] P. Díaz-Vivancos, K. Majourhat, J.A. Fernández, J.A. Hernández, A. Piqueras, J.

Plant Growth Regul. 65 (2011) 119–126.[15] H. Zavaleta-Mancera, H. López-Delgado, H. Loza-Tavera, M. Mora-Herrera, C.

Trevilla-García, M. Vargas-Suárez, H. Ougham, J. Plant Physiol. 164 (2007)1572–1582.

Q. Xu et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 123 (2014) 327–335 335

[16] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, New York,2006.

[17] J.R. Lakowica, G. Weber, Biochemistry 12 (1973) 4161–4170.[18] V.D.W. Marco, S. Lorenzo, J. Mol. Struct. 998 (2011) 144–150.[19] L. Stryer, Annu. Rev. Biochem. 47 (1978) 819–846.[20] R.F. Beers, I.W. Sizer, J. Biol. Chem. 195 (1951) 133–140.[21] A.G. Marangoni, Enzyme Kinetic, A Modern Approach, John Wiley & Sons,

Hoboken, New Jersey, 2002.[22] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J.

Olson, J. Comput. Chem. 19 (1998) 1639–1662.[23] in http://www.pdb.org.[24] M. Bartoszek, W.W. Su, Polish J. of Environ. Stud. 15 (2006) 41–43.[25] G. Zhang, Q. Que, J. Pan, J. Guo, J. Mol. Struct. 881 (2008) 132–138.[26] W.R. Ware, J. Phys. Chem. 66 (1962) 455–458.[27] P.D. Ross, S. Subramanian, Biochemistry 20 (1981) 3096–3102.[28] J.M. Anni, K.A.S. Vanderkooi, T. Yonetani, S.C. Hopkins, L. Herenyi, J. Fidy,

Biochemistry 33 (1994) 3475–3486.[29] Z.X. Chi, R.T. Liu, H. Zhang, Biomacromolecules 11 (2010) 2454–2459.[30] Y.J. Hu, Y. Liu, W. Jiang, R.M. Zhao, S.S. Qu, J. Photochem. Photobiol. B: Biol. 80

(2005) 235–242.[31] H.M. Zhang, J. Chen, Q.H. Zhou, Y.Q. Shi, Y.Q. Wang, J. Mol. Struct. 987 (2011)

7–12.

[32] J. Eisinger, B. Feuer, A.A. Lamola, Biochemistry 8 (1969) 3908–3915.[33] W.Y. He, Y. Li, H.Z. Si, Y.M. Dong, F.L. Sheng, X.J. Yao, Z.D. Hu, J. Photochem.

Photobiol. A 182 (2006) 158–167.[34] N. Beffagna, I. Lutzu, J. Exp. Bot. 58 (2007) 4183–4194.[35] L. Michelet, T. Roach, B.B. Fischer, M. Bedhomme, S.D. Lemaire, A. Krieger-

Liszkay, Plant Cell Environ. 36 (2013) 1204–1213.[36] B. Yang, F. Hao, J. Li, D. Chen, R. Liu, J. Photochem. Photobiol. B 128 (2013) 35–

42.[37] Y. Wei, X. Ye, J. Food Process. Preserv. 35 (2011) 587–595.[38] Z.-M.H. Araceli, L.-D. Humberto, L.-T. Herminia, M.-H. Martha, T.-G. Claudia, V.-

S. Martín, O. Helen, J. Plant Physiol. 164 (2007) 1572–1582.[39] U. Conrath, Z.L. Chen, J.R. Ricigliano, D.F. Klessig, Proc. Natl. Acad. Sci. USA 92

(1995) 7143–7147.[40] A. Fersht, Enzyme Structure and Mechanism., second ed., W.H. Freeman and

Company, New York, 1985.[41] A. Fernandes, G. Gregoriadis, Biochim. Biophys. Acta 1293 (1996) 90–96.[42] B. Sudhamalla, M. Gokara, N. Ahalawat, D.G. Amooru, R. Subramanyam, J. Phys.

Chem. B 114 (2010) 9054–9062.[43] D. Darr, I. Fridovich, Biochem. J. 226 (1985) 781–787.[44] J.A. Poveda, M. Prieto, J.A. Encinar, J.M. González-Ros, C.R. Mateo, Biochemistry

42 (2003) 7124–7132.