Interaction of 7-hydroxyflavone with human serum albumin: A spectroscopic study

Available online at www.sciencedirect.com

www.elsevier.com/locate/jphotobiol

Journal of Photochemistry and Photobiology B: Biology 90 (2008) 33–40

Interaction of 7-hydroxyflavone with human serum albumin:A spectroscopic study

Anwesha Banerjee, Kaushik Basu, Pradeep K. Sengupta *

Biophysics Division, Saha Institute of Nuclear Physics 1/AF, Bidhannagar, Kolkata 700 064, India

Received 21 March 2007; received in revised form 15 October 2007; accepted 28 October 2007Available online 1 November 2007

Abstract

Numerous recent investigations have revealed that various synthetic as well as therapeutically active natural flavonoids possess novelluminescence properties that can serve as highly sensitive monitors for exploring their interactions with relevant physiological targets.Here we report a detailed study on the interactions of the model flavone, 7-hydroxyflavone (7HF) with the plasma protein human serumalbumin (HSA), employing electronic absorption, fluorescence (steady state and time resolved) and induced circular dichroism (ICD)spectroscopy. The spectral data indicate that in the protein matrix, the neutral 7HF molecules are predominantly transformed to a con-jugate anion (7HFA) by a proton abstraction in the ground state. The protein (HSA) environment induces dramatic enhancements in thefluorescence emission intensity, anisotropy (r) and lifetime (s) values, as well as pronounced changes in the fluorescence excitation andemission profiles of the fluorophore. Moreover, evidence for efficient Forster type resonance energy transfer (FRET, from tryptophan to7HFA) is presented, from which we infer that the binding site of 7HF in HSA is proximal (estimated distance, R = 23.6 A) to the uniquetryptophan – 214 residue present in the inter-domain (between IIA and IIIA domains) loop region of the protein. The binding constant(K = 9.44 · 104 M�1) and the Gibbs free energy change (DG = �28.33 kJ/mol) for 7HFA–HSA interaction have been estimated from theemission data. Finally, the near-UV circular dichroism (CD) studies show that the electronic transitions of 7HF are strongly perturbedon binding to the chiral host (HSA), leading to the appearance of ICD bands. Implications of these results are discussed.� 2007 Elsevier B.V. All rights reserved.

Keywords: Protein–flavone interaction; 7-Hydroxyflavone; Steady state fluorescence; Fluorescence lifetime; FRET; Induced circular dichroism (ICD)

1. Introduction

Flavones and related compounds of the flavonoid groupare widely distributed in plants of higher genera [1] wherethey perform many, wide and varied functions of vitalimportance to the survival of plants in the natural environ-ment [2–4]. Recent interest in flavonoids stems from mainlytwo different contexts of intrinsic importance. Firstly,flavonoids are among the best known molecular systemsexhibiting intramolecular excited state proton transfer(ESPT) and dual fluorescence behavior [5,6]. Thus they

1011-1344/$ - see front matter � 2007 Elsevier B.V. All rights reserved.

doi:10.1016/j.jphotobiol.2007.10.005

* Corresponding author. Tel.: +91 033 23370379x3508; fax: +91 03323374637.

E-mail addresses: [email protected], [email protected] (P.K. Sengupta).

can serve as useful models for mechanistic studies on ESPTand related photophysical aspects. Judicious exploitationof the luminescence parameters of flavonoids provideremarkably sensitive means for exploring the structure,function, dynamics, interactions and microenvironmentsin complex biological systems such as proteins. Secondly,such compounds are found to possess novel therapeuticproperties of high potency and low systemic toxicity. Inter-est on this latter aspect dates back to 1936, when Szent-Gyorgyi [7] first drew attention to the therapeuticallybeneficial role of dietary flavonoids. Recent years have wit-nessed revitalized attention in this area with an explosivegrowth of research on various bioactive flavonoids with abroad range of therapeutic properties (effective against can-cers, tumors, allergies, cardiac problems, inflammation,AIDS, etc.) [8–12], making them promising alternatives

mailto:[email protected]

mailto:pradeepsinp@

34 A. Banerjee et al. / Journal of Photochemistry and Photobiology B: Biology 90 (2008) 33–40

to conventional drugs in current use. In this context thequestion of possible target molecules and the mode ofinteractions with targets constitute an important focus ofcurrent pharmacological research on flavonoids.

Several reports have appeared in the literature indicatingthat proteins (including enzymes) are frequently the targetsfor therapeutically active flavonoids of both natural andsynthetic origin [9,13,14]. In this context, studies on theinteraction of flavonoids with serum albumins are particu-larly notable since serum albumins play a critical role in thetransport and disposition of flavonoids [15] therebyincreasing their bioavailibility. While a number of bio-chemical and molecular biological investigations have beenreported in this area, a flurry of recent spectroscopic stud-ies have been especially crucial in enriching our under-standing of the modes of binding of various flavonoids toalbumins [16–23]. However, as structural variations in theflavonoid structure (e.g. position and nature of substitu-ents) have been found to strongly determine their biologi-cal activity, a detailed understanding of their binding andtherapeutic properties could be obtained by studying theinteraction of different structural variants of flavones withserum albumins.

7-Hydroxyflavone (7HF, Scheme 1) is a simple modelfor naturally occurring biologically active flavones exhibit-ing excited-state proton transfer (ESPT) and dual emissionbehaviour [24–34]. It is to be noted that unlike the morewidely studied cases of 3-hydroxyflavone (3HF, a proto-type model flavonol) and its derivatives, where the ESPTis intrinsic (i.e. proceeding across an internal H-bond ofthe molecule) and barrier free [35], the ESPT in 7HF(where proton donor and acceptor sites are not locatedadjacent to each other) is solvent assisted and consequentlystrongly depends on the nature of the solvent medium. Thishas made the photophysics of 7HF a topic of particularinterest. Time resolved fluorescence spectroscopy and tran-sient absorption measurements using two step laser excita-tion (TSLE) [26], indicated that the ESPT of 7HF inmethanol solution involves the formation of two types ofphototautomers in the excited state as well as in the groundstate. The unusual dependence of the luminescence of 7HFon the excitation wavelength (kexc) in alcohol glass at 77 Kwas revealed in a study by Frolov et al. [27]. When kexc was

Scheme 1. 7-hydroxyflavone.

313 nm, a green structured phosphorescence was observedwith a lifetime of �1 ns while a broad blue fluorescence(kmax � 450 nm) appeared when the kexc was changed to365 nm. On increasing the temperature to 298 K, the bluefluorescence shifted to longer wavelengths leading to agreen fluorescence (kmax � 537 nm). Later it was estab-lished [25,28,29] that the blue and green fluorescence weredue to selective excitation of the same ground state species,namely the conjugate anion of 7HF (7HFA), whose fluo-rescence is strongly modulated by solvent relaxation owingto a large change in dipole moments between the groundand excited states of 7HFA.

In previous studies from our laboratory, interesting usesof 7HF as a sensitive fluorescence probe for exploring thelocal environments in membrane mimetic organized assem-blies (normal and reverse micelles) [28–32] were demon-strated. In the present work we have studied theinteraction of 7-hydroxyflavone (7HF) with the circulatoryprotein human serum albumin (HSA) via electronicabsorption, steady-state and time resolved fluorescenceand induced circular dichroism techniques. To our knowl-edge this is the first report on the interaction of 7HF withhuman serum albumin providing critical insights regardingthe nature of binding of 7HF with HSA, in both qualitativeand quantitative detail.

2. Materials and methods

2.1. Materials

7HF was purchased from Aldrich Chemical Companyand was used without further purification. The purity ofthe sample was checked by comparing the electronicabsorption and emission spectra with published data [1].HSA (essentially fatty acid free) and HEPES buffer wereobtained from Sigma. The water used was triple distilled,the final stage being carried out using a quartz distillationsystem.

Stock solutions of the protein (HSA) were made up in50 mM HEPES buffer (pH 7.0) in quartz distilled water.Aliquots from the HSA stock were taken and adequateamounts of HEPES buffer added to attain the desired con-centrations. The concentration of the protein was variedfrom 0 lM to 30 lM in steps of 3 lM, respectively. A con-centrated stock solution of 7HF (3 mM) was prepared inspectrograde methanol (Merck) and small aliquots fromthis stock solution were added to the HSA solutions eachto maintain the flavone concentration at 15 lM (finalmethanol concentration in buffer was <1%).

2.2. Methods

Electronic absorption spectra were recorded with a Cecil7500 spectrophotometer. Steady-state fluorescence mea-surements were carried out with Hitachi (model F-4010)and Perkin–Elmer LS-55 spectrofluorometers, respectively.The fluorescence spectra were corrected for the wavelength

A. Banerjee et al. / Journal of Photochemistry and Photobiology B: Biology 90 (2008) 33–40 35

dependence of the sensitivity of the apparatus. The fluores-cence quantum yields were calculated relative to quininesulphate in 0.1 N H2SO4 (uS = 0.53) using the formula[36],

/F ¼ /S

F X

F S

� �g2

X

g2S

� �AS

AX

� �ð1Þ

where FX and FS are the integrated area under the fluo-rescence emission profiles of the sample and standard solu-tions, respectively, and, AX and AS are the correspondingabsorbance values at the excitation wavelength and gdenotes the refractive index of the solvent medium. Theoptical densities of the sample and reference were matchedat the excitation wavelength and maintained at �0.14.

The binding constant for the protein–flavone complexwas estimated from the emission data (see Section 3). Fluo-rescence titrations were performed on multiple occasionsusing the same experimental conditions. The binding con-stant values obtained from such independent measure-ments were found to be within ±7% of that reported here.

The fluorescence anisotropy (r) values were obtainedusing the following expression,

r ¼ IVV � GIVH

IVV þ 2GIVH

ð2Þ

where IVV and IVH are the vertically and horizontallypolarized components of probe emission with excitationby vertically polarized light at the respective wavelengthand G is the sensitivity factor of the detection system [36].Each intensity value used in this expression represents thecomputer averaged values of 10 successive measurements.

Fluorescence lifetime (s) measurements were performedwith time resolution �1.2 ns and spectral resolution�30 nm using an Edinburgh Instruments nanosecond timecorrelated single photon counting setup with a 370 nmnanosecond diode laser source (IBH, U.K., nanoLED-03). Data analysis was carried out by a deconvolutionmethod using a non-linear least square fitting programmeand fitted with a multi exponential decay function,F(t) =

PiAi exp(�t/si),

PiAi = 1. The goodness of fit

was estimated by using v2 values. The average lifetimesÆsæ were calculated from the following equation,

hsi ¼ A1s1 þ A2s2

A1 þ A2

ð3Þ

where A1, A2 and s1, s2 represent the amplitudes and timeconstants, respectively, of the individual components inbiexponential decay profiles [36,37]. The errors in the life-time values were estimated to be within 1%.

Forster’s theory of resonance energy transfer [38] wasapplied to determine the mean distance between the trypto-phan moiety in HSA and the 7HF binding site. The effi-ciency of energy transfer (E) is related to the distance R

(A) between the donor and acceptor by,

E ¼ R60

R60 þ R6

� �¼ 1� F DA

F D

� �ð4Þ

where FDA and FD are the fluorescence intensities of thedonor in presence and absence of acceptor respectivelyand R0 is the critical Forster distance. R0 is calculated usingthe relation,

R0 ¼ 9:78� 103 j2u0Jg�4� �1=6 ðin AÞ ð5Þ

where J is the spectral overlap integral between the donoremission and acceptor absorption, j2 is the orientation fac-tor assumed equal to 2

3for isotropic donor and acceptor, u0

is the quantum yield of the donor (taken as 0.11 of trypto-phan in HSA [39]) and g is the refractive index of the med-ium (taken as 1.4, as is typically assumed for biomoleculesin aqueous solution [36]). The overlap integral, J, is esti-mated as follows:

JðkÞ ¼R

F ðkÞeðkÞk4dkRF ðkÞdk

ð6Þ

where F(k) and e(k) represent the fluorescence intensity ofthe donor and the molar absorption coefficient of theacceptor respectively, at the wavelength k. The overlap areawas obtained by integrating the overlap region with theOrigin 6.1 software. Using the above parameters the valueof ‘R’ was calculated from Eq. (4).

Circular dichroism (CD) spectra were recorded on aJasco J-600 spectropolarimeter, using a rectangular cuvettewith 1 cm pathlength. The spectra were accumulated threetimes with a bandwidth of 1.0 nm and a scanning step of0.1 nm at a scan speed of 50 nm/min. Induced CD spectrawere obtained as the CD of flavone-HSA mixture minusthe CD of HSA alone. All spectral measurements were car-ried out at ambient temperature (298 K).

3. Results and discussions

3.1. Emission behaviour of 7HF in HSA

Fig. 1a shows typical absorption spectra of 7HF withgradual addition of human serum albumin (HSA). It isobserved that in aqueous buffer, there is a very weak band�365 nm in addition to the characteristic band maxima forthe neutral form of 7HF [28]. We find that addition of HSAleads to the gradual appearance of a pronounced band�365 nm (see Figs. 1a and a (inset)). Previous studies[25,28,33] in homogeneous solvent media, have attributedthe occurrence of such a band to the conjugate anion of7HF (7HFA) formed in the ground state (pKa = 7.39).Fig. 1b represents the absorption behaviour of 7HF inaqueous medium at different pH. Under alkaline condi-tions a pronounced band �365 nm appears which com-pletely diminishes in acidic pH. From the availableliterature data [25,28,33] and these studies it is evident thatunder alkaline conditions, the anionic species of 7HF(7HFA) is formed due to abstraction of proton from the7-hydroxyl moiety. The fact that the 365 nm band isenhanced on addition of protein indicates that the proteinenvironment facilitates the formation of the 7HFA species.

Fig. 2. Fluorescence emission spectra of 7HFA (15 lM) in presence ofvarious concentrations of HSA (0! 30 lM) obtained at kexc = 365 nm.

Fig. 1. (a) Electronic absorption spectra of 7HF (15 lM) in aqueousbuffer containing varying concentrations of HSA (0! 30 lM) at 298 K(Inset shows the absorption spectra of 30 lM HSA (- - - - -), 7HF/aqueousbuffer (ÆÆÆÆ) and 7HF bound to HSA (–)). (b) Electronic absorption spectraof 7HF at pH 3.5 (- - - - -), pH 7.0 (–) and pH 9.5 (ÆÆÆÆ).


Fig. 2 shows the emission spectra of 7HF in presence ofdifferent concentrations of HSA obtained on selective exci-tation of the anion at kexc � 365 nm. In absence of HSA7HFA emits very weakly giving rise to a green emissionband with kem � 524 nm is observed. Interestingly on addi-tion of protein (HSA) the emission profile shows dramaticchanges leading to the appearance of a greatly blue shiftedemission band centered at �488 nm accompanied byenhancement in the fluorescence emission intensity. Theposition of the band maximum remains essentiallyunchanged with increase in HSA concentration while theemission yield (which is almost negligible in aqueous buf-fer) gradually increases until saturation is reached at�30 lM HSA, indicating that more and more 7HF mole-cules bind to HSA. From spectral assignments based onthe literature data [25,28] and the observed absorptionand emission data taken together one can conclude thatthe microenvironment at the binding site in the protein(which presumably contains basic amino acids capable ofproton abstraction) facilitates the creation of the conjugateanion of 7HF (7HFA). The observed blue shift in the emis-sion maximum of 7HFA (from 524 nm in aqueous buffer to488 nm in HSA) can be attributed to restriction on the sol-vent dipolar relaxation imposed by the motionally con-strained environment of the fluorophore in the proteinmatrix. The weak band at � 415 nm arises due to Ramanscattering of the excitation light.

The protein–flavone binding for a 1:1 stoichiometry arerepresented by the following equations,

1

DF¼ 1

DF max

þ 1

Ka½Lt�

� �1

DF max

� �: ð7Þ

and

DF ¼ KaDF max½Lt�1þ Ka½Lt�

ð8Þ

where the former Eq. (7) is the double reciprocal form ofthe nonlinear equation Eq. (8). Here DF = Fx – F0, andFx and F0 represent the fluorescence intensities of 7HF inthe presence and absence of the added total HSA concen-tration ([Lt]), respectively and DFmax is the maximumchange in fluorescence intensity [16]. On substituting thefluorescence data for the anionic species of 7HF in Eq.(7) (see Fig. 3a) it is found that the graph exhibits good lin-earity which confirms that the 7HFA–HSA binding stoichi-ometry is distinctly 1:1. The binding parameters estimatedfrom this linear double reciprocal plot were used as an ini-tial guess in the iterative NLR fit using Eq. (8), because anonlinear least-squares regression (NLR) analysis is knownto give more accurate values of the binding constants thanthe corresponding linearized form (see Eq. (7)) [40]. The Ka

value estimated from the NLR fit (Fig. 3b) for the 7HFA–HSA complex comes out to be (9.44 ± 0.41) · 104 M�1, theerror being associated with the fitting procedure. The stan-dard Gibbs free energy change (DG0 = �RTlnK) comes outto be approximately �28.33 kJ/mol at 25 �C which indi-cates the spontaneity of the binding of 7HFA with HSA.

Fig. 3. Binding constant plots obtained from the fluorescence intensitiesrecorded for the 7HFA–HSA complexes (kex = 365 nm, kem = 490 nm).(a) Double reciprocal plot of 1/D F versus 1/[Lt]. (b) The result of curve fitobtained by a nonlinear least-squares regression (NLR) analysis of theemission data.

Fig. 4. (a) Variation in the fluorescence anisotropy (r) of 7HFA (15 lM)with increasing protein (HSA) concentration (kex = 365 nm, kem = 490 nm).(b) Polarization excitation spectrum of 7HFA showing the dependence ofanisotropy on the excitation wavelength (kex) (kem = 490 nm).


3.2. Fluorescence anisotropy studies

Fluorescence anisotropy (r) measurements were alsoperformed since this parameter serves as a sensitive indica-tor for monitoring ligand binding to macromolecular sys-tems [16–18,22], biomembranes [41], etc. The anisotropyvalues of fluorophores are very low in fluid solution wherethe fluorophore molecules can freely rotate, and increase inmotionally constrained environments [36]. Fig. 4a presentstypical fluorescence anisotropy data obtained for 7HFA(kexc = 365 nm). The plot shows the pronounced increase

Table 1Photophysical parameters of 7HF in HSA and methanol (kexc = 370 nm)

Medium kemmax (nm) r s1 (ns) A1

MeOH 520 0.03 0.48 0.997HSA 480 0.311 0.97 0.696

HSA, human serum albumin; [HSA] = 30 lM; [7HF] = 15 lM.

in the anisotropy (r) values upon addition of HSA. Thehigh values of anisotropy obtained at even low concentra-tions of the protein indicate that in the protein bound form,the 7HF molecule exists mainly in the anionic state. More-over appreciably high anisotropy values indicate that7HFA molecules are ‘locked’ in motionally restricted sitesof HSA.

Fig. 4b depicts the polarization excitation spectrum of7HFA. It is evident that the anisotropy reaches a largevalue (0.31) for excitations at the red-edge of the absorp-tion band (where 7HFA absorbs maximally). On decreas-

s2 (ns) A2 Æsæ (ns) v2 uF

2.37 0.003 0.48 1.319 0.00192.8 0.304 1.53 1.184 0.023

Fig. 5. (a) The fluorescence excitation spectrum of 7HF in the presence ofHSA (30 lM) monitored at kem = 490 nm showing occurrence of FRET(tryptophan of HSA! 7HF). (b) Spectra showing the overlap between theemission (- - - - -) of the donor (Trp) and acceptor (7HF) absorption (—).


ing the excitation wavelength the anisotropy decreasesdrastically. This confirms the binding of only one speciesof 7HF, i.e. the anionic form to HSA.

3.3. Fluorescence lifetime studies

We have also carried out fluorescence lifetime studies of7HFA in the protein matrix. The decay parameters, aver-age fluorescence lifetimes Æsæ and v2 values are listed inTable 1. Interestingly the decay parameters are greatlyaffected by the protein environment, with the average life-time Æsæ changing from 480 ps recorded in methanol to1.526 ns in HSA. (N.B.: Due to extremely low emissionyield of the flavone in aqueous medium, the results in pro-tein have been compared with those in methanol chosen asa typical homogeneous medium, for reference). In the pres-ent study, the fluorescence decay time of the fluorophore isfound to be lower than the pulse full width at half maxi-mum (FWHM, �1.2 ns) of the diode laser excitationsource, and therefore the decay times could not be obtaineddirectly. Instead the method of deconvolution using non-linear least square fitting programme, provided with thedata analysis software, was used to determine the lifetimevalues [37]. The enhanced lifetime is attributable to thereduction in the non-radiative decay rates in the rigid pro-tein matrix, which could imply that deactivation due tointernal conversion (IC) processes [42] are reduced and thatthe probe molecule is bound to a rigid pocket in the HSAmatrix where conformational changes (especially related tophenyl ring torsion) which occur during the excited statelifetime are considerably diminished.

3.4. Fluorescence resonance energy transfer

We then sought evidence for fluorescence resonanceenergy transfer (FRET) which serves as a reliable indicatorof the proximity of donor and acceptor sites in a macro-molecule [36]. The excitation profile monitored at�490 nm (Fig. 5a) reveals that besides the band(s) corre-sponding to the absorption of 7HF, a band withkex � 290 nm occurs, which is attributable to tryptophan(trp) absorption. This is indicative of Forster type fluores-cence resonance energy transfer (FRET) from the singletrp-214 moiety of HSA.

Fig. 5b shows the spectral overlap of the donor (trp)emission and the acceptor (7HF) absorption. The magni-tude of the overlap integral (J) estimated from Eq. (6)comes out to be 0.79 · 10�15 M�1 cm3. Substituting thisparameter in Eq. (5), the critical Forster distance (R0)was estimated as 23.5 A. From the values of the donor fluo-rescence intensity measured in the absence and presence of7HF, the energy transfer efficiency (E) was calculated as0.495. Thus using these values obtained for the efficiencyand R0, the distance ‘R’ was estimated to be 23.6 Abetween the 7HFA binding site and trp-214 in HSA. Thesevalues are very similar to those obtained for quercetin inHSA [43].

From the fluorescence data presented in this work it canbe inferred that 7HF binds to HSA predominantly at a sitelying in close proximity to trp-214 situated in the interdo-main (between IIIA and IIA domains) cleft region of theprotein [17–19,22,44]. This site is most likely to be closeto an amino acid having a charged side chain [17–19], facil-itating the formation of the anionic species of 7HF(7HFA), which is the predominant form in which the fluo-rophore exists in the protein matrix.

3.5. Induced circular dichroism studies

Induced circular dichroism (ICD) spectroscopy providescrucial and discriminating evidences for exploring the bind-ing of achiral ligands to chiral hosts [19,45,46]. As flavonesare optically inactive in nature, their binding to a chiral

Fig. 6. Induced CD spectra of 7HF (50 lM) in presence of HSA (30 lM).The corresponding UV absorption spectra of 7HF in methanol (- - - - -) andin presence of protein (—) is also shown.

Table 2Wavelength maxima (k) and ellipticity (e) values of the induced CD bandsof 7HF (50 lM) bound to HSA (30 lM)

Absorption region k (nm) e (mdeg)

Band I 378 �0.89349 +0.67330 +2.41318 +3.04307 +1.95

Band II 286 +1.32272 +4.78253 +7.35


host like HSA is expected to lead to the appearance ofinduced chirality. Zsila et al. [19] have characterized thebinding of the flavonol quercetin to HSA where threeICD bands were observed in the absorption region of quer-cetin. Fig. 6 shows the ICD spectrum of 7HF on binding toHSA. The UV absorption spectra of 7HF in methanol andin presence of HSA are also included for reference. It isfound that HSA induces prominent CD bands in the regionfrom 250 to 400 nm. The UV-absorption spectrum of flav-ones is characterized by two principal absorption bandsdesignated as Band I (300–380 nm) and Band II (240–285 nm). Band I is attributed to arise from light absorptionfrom the cinnamoylic portion (B+C) portion of the mole-cule, and Band II from the benzoyl moiety (A+C) ring ofthe flavonoid molecule. These bands are believed to beessentially p–p* in nature [24]. The UV spectrum of 7HFin methanol is characterized by the characteristic bandmaxima for the neutral form of 7HF around 310 nm(Band I) accompanied by two other bands around�252 nm and �270 nm in the Band II region. In addition,in the presence of the protein there is a significant contribu-tion around 365 nm due to enhanced formation of the con-jugate anion (which is present only in very small amounts

in absence of added protein), as noted from UV absorptionand fluorescence studies described earlier. Interestinglyprominent positive induced CD bands are noted in theBand I and Band II electronic absorption regions of the fla-vone (Table 2). Of the eight ICD bands, five (�378, 307,286, 272, and 253 nm) can be correlated with correspond-ing bands in the electronic absorption spectrum, whilethree of them (349, 330, and 318 nm) could not be assignedat this stage. The band around 370 nm is believed to arisedue to the anionic form of 7HF.

Thus the appearance of induced CD bands correspond-ing to electronic transitions observed in the absorptionspectrum shows that such transitions are strongly per-turbed chirally in the binding site of the protein.

3.6. Concluding remarks

The research described herein provides insights high-lighting the photophysical properties of 7HF in a typicalprotein environment. Furthermore it is evident that 7HFcan serve as a highly sensitive extrinsic fluorescence probefor exploring protein–flavone interaction and characteriza-tion of possible binding site(s). Moreover circular dichro-ism studies reveal that binding of 7HF, an opticallyinactive molecule, to chiral binding sites in the protein, per-turbs the electronic transitions of the flavone asymmetri-cally, leading to the appearance of induced circulardichroism (ICD) bands.

Acknowledgements

A.B. thanks CSIR, India for a Senior Research fellow-ship (CSIR Grant award no. 9/489 (44)/ 2002- EMR-I).We thank Prof. N. Chattopadhyay, Jadavpur Universityfor access to the fluorescence lifetime setup in his labora-tory and his students (A. Chakraborty and Dr. A. Mallik)for their kind help. We also acknowledge the kind assis-tance of Mr. J. Guin for the Circular Dichroism studiesperformed at the P.D. Laboratory, Bose Institute.

References

[1] T.J. Mabry, K.R. Markham, M.B. Thomas, The Systematic Identi-fication of Flavonoids, Springer Verlag, New York, Heidelberg,Berlin, 1970.


[2] R. Brouillard, A. Cheminat, in: V. Cody, E. Middleton, J.B.Harborne, A. Beretz (Eds.), Plant flavonoids in biology and medicineII: biochemical, Cellular and Medicinal Properties, Alan R. Liss, NewYork, 1988, p. 93.

[3] M. Tevini, A.H. Teramura, Photochem. Photobiol. 50 (1989) 479–487.

[4] R.F. Fischer, S.R. Long, Nature 357 (1992) 655–660.[5] P.K. Sengupta, M. Kasha, Chem. Phys. Lett. 68 (1979) 382–385.[6] A.P. Demchenko, S. Ercelen, A.D. Roshal, A.S. Klymchenko, Polish

J. Chem. 76 (2002) 1287–1299.[7] St. Rusznyak, A. Szent-Gyorgi, Nature 138 (1936) 27–29.[8] J.B. Harborne, in: V. Cody, E. Middleton, J.B. Harborne, A. Beretz

(Eds.), Biology and medicine II: biochemical, Cellular and MedicinalProperties, Alan R. Liss, New York, 1988, p. 17.

[9] S.W. Lamson, M.S. Brignall, Altern. Med. Rev. 5 (2000) 196–208.[10] P.K. Sengupta, A. Banerjee, B. Sengupta, in: Kaneco, S. (Ed.), Photo/

Electrochemistry and Photobiology in the Environment, Energy andFuel, Research Signpost, Kerala, India, 2006, pp. 207–226 andreferences cited therein.

[11] C.D. Kanakis, P.A. Tarantilis, M.G. Polissiou, S. Diamantoglou,H.A. Tajmir-Riahi, J. Mol. Struct. 798 (2006) 69–74.

[12] O.M. Anderson, K.R. Markham (Eds.), Flavonoids: Chemistry Bio-chemistry and Applications, CRC Press, Boca Raton, Florida, 2006.

[13] P.C. Ferricola, V. Cody, E. Middleton, Biochem. Pharmacol. 38(1989) 1617–1624.

[14] R.S. Brinkworth, M.J. Stoermer, D.P. Fairlie, Biochem. Biophys.Res. Commun. 188 (1992) 631–637.

[15] O. Dangles, C. Dufour, C. Manach, C. Remsey, Methods Enzymol.335 (2001) 319–333.

[16] J. Guharay, B. Sengupta, P.K. Sengupta, Proteins: Struct., FunctionGenet. 43 (2001) 75–81.

[17] B. Sengupta, P.K. Sengupta, Biochem. Biophys. Res. Commun. 299(2002) 400–403.

[18] B. Sengupta, P.K. Sengupta, Biopolymers 72 (2003) 427–434.[19] F. Zsila, Z. Bikadi, M. Simonyi, Biochem. Pharmacol. 65 (2003) 447–

456.[20] J. Tian, J. Liu, J. Xie, X. Yao, Z. Hu, X.J. Chen, Photochem.

Photobiol. B: Biol. 74 (2004) 39–45.[21] Q. Bian, J. Liu, J. Tian, Z. Hu, Int. J. Biol. Macromol. 34 (2004) 275–

279.[22] B. Sengupta, A. Banerjee, P.K. Sengupta, J. Photochem. Photobiol.

B: Biol. 80 (2005) 79–86.[23] W. He, Y. Li, C. Xue, Z. Hu, X. Chen, F. Sheng, Bioorg. Med. Chem.

13 (2005) 1837–1845.

[24] R. Schipfer, O.S. Wolfbeis, A. Kneirzinger, J. Chem. Soc. PerkinTrans. II (1989) 1443–1448.

[25] O.S. Wolfbeis, M. Leiner, P. Hochmuth, H. Geiger, Ber. Bunsenges.Phys. Chem. 88 (1984) 759–767.

[26] M. Itoh, T. Adachi, J. Am. Chem. Soc. 106 (1984) 4320–4324.[27] Y.L. Frolov, Y.M. Sapozhnikov, S.S. Barer, N.N. Pogodaeva, NA.

Tyukavkina, Russ. Chem. Bull. 23 (1974) 2279–2281.[28] M. Sarkar, P.K. Sengupta, J. Photochem. Photobiol. A: Chem. 48

(1989) 175–183.[29] M. Sarkar, P.K. Sengupta, Ind. J. Phys. 65B (5) (1991) 489–494.[30] R. Chaudhari, J. Guharay, P.K. Sengupta, J. Photochem. Photobiol.

A: Chem. 101 (1996) 241–244.[31] M. Sarkar, J. Guharay, P.K. Sengupta, J. Photochem. Photobiol. A:

Chem 95 (1996) 157–160.[32] B. Sengupta, J. Guharay, P.K. Sengupta, J. Surf. Sci. Technol. 14

(1–4) (1998) 150–156.[33] O.S. Wolfbeis, R. Schipfer, Photochem. Photobiol. 34 (1981) 567–

571.[34] K.R. Markham, T.J. Mabry, in: J.B. Harborne, H. Mabry, T.J.

Mabry (Eds.), The Flavonoids, Chapman and Hall, London, 1975,pp. 44–77.

[35] M. Kasha, Acta. Phys. Pol. A. 71 (1987) 717–729.[36] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, third ed.,

Springer, New York, 2006.[37] D.J.S. Birch, R.E. Imhof, in: J.R. Lakowicz (Ed.), Topics in

Fluorescence Spectroscopy, vol. 1, Plenum Press, New York, 1991,pp. 1–182.

[38] Th. Forster, Ann. Phys. 2 (1948) 55–75.[39] A. Sytnik, I. Litvinyuk, Proc. Natl. Sci. Acad. USA 93 (1996) 12959–

12963.[40] K.A. Connors, Binding Constants: The Measurement of Molecular

Complex Stability, Wiley, New York, 1987.[41] B. Sengupta, A. Banerjee, P.K. Sengupta, FEBS Lett. 570 (2004) 77–

81, and references cited therein.[42] G.J. Smith, K.R. Markham, J. Photochem. Photobiol. A: Chem. 99

(1996) 97–101.[43] B. Mishra, A. Barik, K.I. Priyadarshini, H. Harimohan, J. Chem. Sci.

117 (6) (2005) 641–647.[44] D.C. Carter, X.M. He, Nature 358 (1992) 209–215.[45] F.A. Drake, Circular dichroism, in: S.E. Harding, B.Z. Chowdhry

(Eds.), Protein Ligand Interactions: Structure and Spectroscopy,Oxford University Press, 2001, p. 123.

[46] M. Dockal, D.C. Carter, F. R}uker, J. Biol. Chem. 274 (41) (1999)29303.

Documents

Interaction of 7-hydroxyflavone with human serum albumin: A spectroscopic study