Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45

Contents lists available at SciVerse ScienceDirect

Journal of Photochemistry and Photobiology B: Biology

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

Fluorescence modulation and associative behavior of lumazinein hydrophobic domain of micelles and bovine serum albumin

N. Shaemningwar Moyon, Mullah Muhaiminul Islam, Smritakshi Phukan, Sivaprasad Mitra ⇑Department of Chemistry, North-Eastern Hill University, Shillong 793 022, India

a r t i c l e i n f o

Article history:Received 16 November 2012Received in revised form 13 February 2013Accepted 13 February 2013Available online 26 February 2013

Keywords:LumazineSurfactantBovine serum albuminFluorescence modulationAssociation constantHydrophobic interaction

1011-1344/$ - see front matter � 2013 Elsevier B.V. Ahttp://dx.doi.org/10.1016/j.jphotobiol.2013.02.008

⇑ Corresponding author. Tel.: +91 364 2722634; faxE-mail addresses: smitra@nehu.ac.in, sivaprasadm@

a b s t r a c t

The photophysical behavior of the deprotonated form of lumazine (Lum-anion) was studied in biologi-cally relevant surfactant systems like sodium dodecyl sulfate (SDS), cetyltrimethylammonium bromide(CTAB) and TritonX-100 (TX-100) and also model water soluble protein, bovine serum albumin (BSA),using steady-state and time-resolved fluorescence spectroscopy in buffer solution of pH 12.0. The asso-ciation constant values were calculated from modulated fluorescence intensity of Lum-anion in differentmedium. The interaction of non-ionic surfactant TX-100 was found to be about 10 times greater than SDSand CTAB. However, while the driving force of binding in SDS and/or TX-100 is mainly hydrophobic innature, electrostatic interaction with the oppositely charged micellar head group is the predominant fac-tor in CTAB. The thermodynamic parameters like enthalpy (DH) and entropy (DS) change, etc., corre-sponding to the binding of Lum-anion with BSA were estimated by performing the fluorescencetitration experiment at different temperatures. Thermodynamically favorable and strong binding ofLum-anion (K � 104 M�1) into BSA is due to hydrophobic interaction in the ligand binding domain II.However, the binding mechanism is entirely different in presence of protein denaturing agent like ureaand electrostatic interaction plays a major role under this condition.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Pteridines in their multiple forms are widespread in biologicalsystems and represent an important class of heterocyclic com-pounds with different roles ranging from pigments to cofactors fornumerous redox and one-carbon transfer reactions [1,2]. Lumazines(pteridine-2,4(1,3H)-dione) are natural products from the metabolicdegradation of pterins [3]. The function and properties of the lum-azine protein isolated from blue light emitting bioluminescent pho-tobacteria is similar to that of yellow fluorescence protein (YFP) [4].The protein family consisting of lumazine synthases are known toform several quaternary structures and recently, been engineeredfor versatile applications ranging from catalysis of retro-aldol reac-tion [5], novel molecular encapsulation systems, etc. [6–8]. Lum-azine derivatives are present in cells, since 6,7-dimethyl-8-ribityllumazine is the biosynthetic precursor of riboflavin (vitaminB2). Riboflavin is itself the precursor of flavin mononucleotide(FMN) and flavin adenine dinucleotide (FAD), essential cofactorsfor a wide variety of redox enzymes [9,10]. In a recent report, theuse of lumazine was also demonstrated as a methanogen inhibitor[11]. The participation of pterins in photobiological processes hasbeen suggested and/or demonstrated in the past decades and stillcontinues to a topic of active research interest.

ll rights reserved.

: +91 364 2550076.yahoo.com (S. Mitra).

Lumazine presents different acid–base behavior in aqueoussolutions. The only relevant equilibrium at physiological pH in-volves the neutral (acid) form and the monoanion (basic) form(Fig. 1), with a pKa value of 8.0 ± 0.1 [12,13]. In recent communica-tion, we have described the fluorescence behavior of neutral lum-azine (Lum) with a combination fluorescence experiment and highlevel density functional theory (DFT) calculation [14]. Also, thephotophysical behavior of the luminol (5-amino-2,3-dihydro-1,4-phthalazinedione) and lumichrome (7,8-dimethylalloxazine), theheterocyclic systems containing similar structural motif to thatin lumazine were reported in homogeneous as well as severalbio-mimicking micro-heterogeneous media [15–18]. Whereas thephotophysics and photochemistry of pterins have been studied indetail [19], little is known about the photochemical behavior oflumazines [20,21]; particularly, in biologically relevant heteroge-neous media like micelles and/or proteins. The relatively high fluo-rescence quantum yield and long lifetime of 6,7-dimethyl-8-ribityllumazine, the prosthetic group of lumazine protein, makesit an attractive probe to study ligand–protein and protein–proteininteraction and also protein dynamics using thermodynamicallyideal, diluted protein solutions [22,23]. In the present paper, we re-port the results on the spectral modulation and binding behavior ofmono-anionic lumazine (Lum-anion) in several surfactant mediaas well as in presence of model water soluble globular protein likebovine serum albumin (BSA) by steady state and time-resolvedfluorescence techniques.

Fig. 1. Steady state spectral absorption (a), fluorescence emission (b) and excitation (c) spectra of �5.0 � 10�6 mol dm�3 lumazine solution in neutral (Lum) anddeprotonated (Lum-anion) form. The excitation wavelengths are 325 and 365 nm, respectively; whereas the excitation spectra were taken by monitoring the emissionwavelength at 455 nm in both the cases.

38 N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45

2. Materials and methods

2.1. Chemicals

The neutral lumazine (Lum) was received from Sigma–AldrichChemical Pvt. Ltd. (product no. L-3307) and used without any furtherpurification. The analytical grade type – II water was obtained fromElix 10 water purification system (Millipore India Pvt. Ltd.). All theexperiments were carried out at ambient temperature of 293 ± 1 Kin buffer solution of pH 12.0 obtained by mixing 100 ml of 0.05 MNa2HPO4 with 53.8 ml of 0.1 M NaOH and the volume was adjustedto 200 ml. The analytical grade reagents were obtained from SiscoResearch Laboratory (SRL), India. As discussed in the previous sec-tion, only the deprotonated form of lumazine (Lum-anion) existsin the solution mixture at this working pH. The surfactants sodiumdodecyl sulfate (SDS, product no. 86201-0), cetyltrimethylammo-nium bromide (CTAB, product no. 85582-0), and TritonX-100 (TX-100, product no. T8787) were procured from Sigma–Aldrich (India)and all were used as received. Essentially fatty acid and globulin free,P99% (agarose gel electrophoresis), lyophilized powder form of bo-vine serum albumin (BSA, Sigma, cat. no. B4287) was also used as re-ceived. The stock solution of the surfactants/protein was preparedby mixing required amount of the substrate in buffer. The final solu-tion pH was controlled with Systronics l-pH system 361 and ad-justed for slight pH variation due to the mixing of the substrates, ifany, by adding required amount of HCl or NaOH. The chromophoreconcentration (�5 lM) was very low to avoid any aggregation andkept constant during spectral measurements; whereas, the finalconcentration of the heterogeneous media was adjusted by addingthe required volume from the stock solution. All the solutions wereprepared afresh and kept for 30 min for settling before the spectro-scopic measurement.

2.2. Experimental procedure

Steady-state absorption spectra were recorded on a Perkin–El-mer model Lambda25 absorption spectrophotometer. Fluorescencespectra were taken in a Hitachi model FL4500 spectrofluorimeterand all the spectra were corrected for the instrument responsefunction. Quartz cuvettes of 10 mm optical path length receivedfrom PerkinElmer, USA (part no. B0831009) and Hellma, Germany(type 111-QS) were used for measuring absorption and fluores-cence spectra, respectively. For fluorescence emission, the samplewas excited at 365 nm unless otherwise mentioned, whereas

excitation spectra were obtained by monitoring at the respectiveemission maximum. In all cases, 5 nm bandpass was used in theexcitation and emission side. Any possible contribution of inner fil-ter effect due to attenuation of the incident light by the quencher isnegligible, since none of the components providing the heteroge-neous environments have any significant absorption at the excita-tion wavelength. Nevertheless, the observed fluorescence intensity(Fobs) was corrected for any possible attenuation of excitationintensity in presence of surfactant and protein medium by usingthe following equation [24,25]:

FðkE; kFÞ ¼ FobsðkE; kFÞ �AðkEÞ

AtotðkEÞð1Þ

where A represents the absorbance of free Lum-anion and Atot is thetotal absorbance of the solution at the excitation wavelength (kE).All steady-state data obtained from at least three separate experi-ments were averaged and further analyzed using Origin 6.0 (Micro-cal Software, Inc., USA).

Fluorescence quantum yields /if

� �were calculated by compar-

ing the total fluorescence intensity under the whole corrected fluo-rescence spectral range with that of a standard (quinine bisulfatein 0.5 M H2SO4 solution, /s

f ¼ 0:546 [26]) with the following equa-tion using adequate correction for solution absorbance (A) and sol-vent refractive index (n) [27].

/if ¼ /s

f �Fi

Fs :1� 10�As

1� 10�Ai �ni

ns

� �2

ð2Þ

The relative experimental error of the measured quantum yieldwas estimated within ±10%. The temperature variation experi-ments were carried out by attaching a circulatory thermostat bath(MLW, Germany, type U2C) to the cell holder.

Fluorescence decay analysis were performed in a LED basedfluorescence lifetime spectrometer (Photon Technology Interna-tional) equipped with a stroboscopic detector. Fluorescence life-time measurement using the stroboscopic method is becomingincreasingly popular [28–30] over the standard time correlatedsingle photon counting (TCSPC) and/or phase modulation tech-niques due to its fast, low cost and easy operating condition, with-out compromising the accuracy [31]. The kinetic traces were fittedwith a sum of exponential decay function by a reconvolution pro-cedure with the lamp pulse profile obtained from scattering of acolloidal suspension of coffee diary whitener using Felix32 soft-ware supplied by PTI.

N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45 39

3. Results and discussion

3.1. Fluorescence behavior in homogeneous solution

The absorption spectrum of lumazine in aqueous solutionshows a strong and broad absorption band within 280–360 nmspectral range with a maximum at 325 nm region. Excitation atthis peak gives dual emission peaks; one very weak band at375 nm and the other quite strong as well as broad band with apeak position at 460 nm. The excitation spectrum correspondingto the 375 nm emission matches exactly with the absorption band;however, the 460 nm emission gives a new excitation band with apeak position at 365 nm. Interestingly, in aqueous buffer solutionof pH 12.0, the absorption and emission peak appears only at365 and 460 nm, respectively (Fig. 1). In accordance with the liter-ature reports, the experimental observations are consistent withdissociation equilibrium of the neutral (Lum) and deprotonated(Lum-anion) lumazine structure corresponding to the absorptionpeak at 325 and 365 nm with the respective emission peaks at375 and 460 nm. Interestingly, in water solution, lumazine mainlyexists in the protonated form (Lum) in the ground state corre-sponding to a single absorption band; however in the excited state,the dissociation equilibrium is mostly shifted toward the deproto-nated form resulting a dual emission band. On the other hand, instrongly alkaline buffer of pH 12.0, the deprotonated structure(Lum-anion) can be considered as the only species both the groundand excited states. The photophysical behavior of lumazine is sum-marized Scheme 1. With reference to the results of Klein and Tati-scheff [32] and also the recent studies [13,33], the deprotonatedspecies can be considered as the mono-anionic form lumazine atthis pH.

In a recent paper [14], we have presented detail theoretical cal-culation result on the existence and possible inter-conversion path-way among different tautomeric forms of lumazine both in theground and excited states. Analyses of the frontier orbitals revealthat the N3 and N1 sites are mostly pertinent for proton dissocia-tion in the ground and excited states, respectively. This result is alsoin good agreement with Klein and Tatischeff as they reported thatthe absorption spectrum of mono-anionic lumazine resemblesstrongly with that of 1-methyl-lumazine (with possibility of proton

Scheme 1. Proton dissociation equilibrium and photophysical behavior of neutral (Lum)

dissociation only at N3); whereas, the emitting properties beingprincipally due to the N1 deprotonated species [32].

Time-resolved fluorescence property of Lum-anion at pH 12.0shows single exponential decay behavior with corresponding de-cay time of 4.6 ± 0.2 ns, when measured with 365 nm LED excita-tion and the emission monitored at 455 nm. Recently, Presiadoet al. [34] reported the fluorescence decay measurement of lum-azine at different solution pH. The single exponential decay behav-ior of Lum-anion reported by these authors is in agreement withour result, although the reported lifetime (7.5 ns) differs signifi-cantly. This might be due to the difference in experimental condi-tion in the two cases. For example, the excitation and emissionwavelengths were set at 365 and 455 nm, respectively in the pres-ent investigation in contrast with 260–300 nm excitation and510 nm emission wavelengths by Presiado et al. [34]. As the ex-cited state photophysical behavior of lumazine is a complex sub-ject with the involvement of several tautomeric species andcorresponding anions, we chose the excitation and emission wave-lengths in matching agreement with the steady state results dis-cussed above. Furthermore, in a recent comment [13], Denofrioet al. pointed out that the pH of the working solution in Ref. [34]was not measured properly and this might also be a reason forthe observed difference in the reported fluorescence lifetime.

3.2. Spectral properties in presence of surfactants

The steady state and time-resolved fluorescence properties ofLum-anion was studied in three model surfactants; namely SDS,CTAB and TX-100, containing anionic, cationic and neutral headgroups respectively. The absorption spectral profile of Lum-anionis remarkably insensitive to the increasing surfactant concentra-tion; however, the fluorescence intensity shows a drastic changein presence of surfactants. Fig. 2 shows the variation of Lum-anionfluorescence intensity with increasing concentration of SDS andCTAB along with the intensity variation pattern shown in the inset.Fig. S1a in the Supplementary section shows few representativeabsorption spectral profiles of Lum-anion in presence of differentconcentrations of TX-100 and SDS; while, the corresponding figurefor fluorescence intensity variation in TX-100 is shown in Fig. S1b.The fluorescence properties of Lum-anion in homogeneous buffer

and mono-anionic (Lum-anion) lumazine in the ground as well as in excited states.

Fig. 2. Variation in Lum-anion fluorescence spectral profile (kexc = 365 nm) with increasing concentration of SDS (a) and CTAB (b). The concentration of surfactants are (a)[SDS]/mM = 0.0 (1), 1.1 (2), 2.7 (3), 4.4 (4), 8.0 (5), 10.1 (6) and 14.6 (7); (b) [CTAB]/mM = 0.0 (1), 0.6 (2), 4.0 (3), 6.7 (4) and 10.2 (5). Inset shows the change in emissionintensity at 455 nm in each case.

Table 1Steady state spectral properties 5 lM Lum-anion solution in homogeneous buffer and in presence of different surfactants.

Medium kabs (nm) kem (nm) /f sf (ns) jr/107 (s�1) jnr/108 (s�1)

Buffer (pH = 12.0) 365 455 0.23 4.6 5.0 1.7SDS (17 mM) 365 455 0.30 5.0 6.2 1.4CTAB (6 mM) 365 455 0.20 4.7 4.6 1.7TX-100 (0.6 mM) 365 455 0.41 6.3 6.4 0.9

Abbreviations used: k = absorption and emission wavelengths; /f = fluorescence quantum yield; sf = fluorescence decay time; jr and jnr represent the radiative and non-radiative decay rate constants calculated from Eq. (5). The concentration of surfactants is mentioned in the parenthesis.

40 N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45

solution as well as in fully micellized condition are listed in Table 1.Apparently, the fluorescence peak position as well as the spectralshape also is not too sensitive towards the micellar medium. Theinvariant nature of the fluorescence emission spectral shape ofLum-anion in SDS, CTAB, and Triton X-100 are similar to the spec-trum shape in homogeneous buffer, suggesting that the probeshares a similar bound structure to the solvated form. It is interest-ing to note that non-ionic TX-100 and anionic SDS provide signifi-cant enhancement in Lum-anion fluorescence intensity. On theother hand, with addition of cationic surfactant CTAB, the fluores-cence intensity initially increases up to certain concentration closeto the critical micelle concentration (cmc � 0.8 mM [35]), followedby a rapid fall with further surfactant addition. It is already wellknown that the initial increase in absorption/fluorescence intensityat very low surfactant concentration is due to the increased solu-bility of the probe. Considerable change of Lum-anion fluorescenceintensity in the surfactant medium beyond cmc indicates a stronginteraction of the probe with the micelle and binding of Lum-anionin the interfacial layer of the respective micellar structures. Consid-ering the anionic nature of the fluorophore, dipole–dipole interac-tion between the probes with CTAB (containing cationicammonium head group) is expected to play a dominant role overthe hydrophobic interaction. As a result of the formation of non-fluorescent or very weakly fluorescent charge neutralized complexdue to the Coulombic interaction, the fluorescence intensity of theLum-anion decreases substantially in CTAB micelles. However, innon-ionic TX-100 and anionic SDS, the hydrophobic interactionprevails and the Lum-anion fluorescence intensity increases dueto substantial reduction in the total non-radiative rate under thearrested condition inside the micellar sub-domain. The differentnature of Lum-anion binding to CTAB and SDS/TX-100 hypothe-sized here is further supported from the calculation of binding con-stant values and also the measurement of fluorescence decay timeunder varying surfactant concentration (see below).

3.2.1. Analysis of binding behavior with the surfactantsThe strength of the interaction between the substrate (S) and

investigated detergents (D) was estimated using the method devel-oped by Hirose and Sepulveda [36] and successfully applied to dif-ferent fluorescent systems [37,38] considering the binding of aprobe by the following equilibrium:

Sa þ D$ Sm ð3Þ

where Sa and Sm denote the substrate concentrations expressed asmolarities in terms of total volume of solution in the aqueous phaseand in the micellar pseudo-phase. Under the condition of [Sm] -� [Dt] and [Dt]� cmc (Dt is the total detergent concentration),the association constant (KS) of the above equilibrium can be relatedwith the fluorescence intensity as follows:

F � F0

½Dt �¼ KS � ðFm � FÞ ð4Þ

where F, F0 and Fm are the total area under the whole fluorescenceemission spectra of the probe in surfactant, water and in fully mic-ellized condition, respectively. Therefore, a plot of (F � F0)/(Fm � F)vs [Dt] gives a straight line, the slope of which gives the value ofthe binding constant, Ks. Some typical plots are shown in Fig. 3and the results are given in Table 2.

It is seen from Table 2 that the binding of Lum-anion in all thesethree surfactant systems are spontaneous with negative Gibbs freeenergy change; however, the extent of interaction is stronger withTX-100 than both in SDS and/or CTAB. It is well known that themicellar size of SDS is quite small with an average radius,R � 30 Å in comparison with CTAB (R � 51 Å) and TX-100(R � 50 Å) [39–41]. On the other hand, the tail length of TX-100(�35.3 Å) is significantly longer than both of SDS and TX-100(�15 and �20 Å, respectively) [42]. The higher value of Lum-anionbinding constant in TX-100 can be explained on the basis of a com-bined effect of these two factors.

Fig. 3. Plot of (F � F0)/(Fm � F) against surfactant concentration. The linear regres-sion of the experimental data with Eq. (4) is shown by the solid line along with thecorrelation coefficient in each case.

Table 2Binding parameters of lumazine anion with different surfactants in aqueous buffer ofpH 12.0.a

Surfactant KS/103 (M�1) DG (kJ mol�1)

SDS 0.325 ± 0.03 �14.09TX-100 2.509±0.24 �19.07CTAB 0.17±0.02 �12.51

a KS values are mean ± standard deviation; the corresponding values of change inGibbs free energy, (DG = �RT ln KS) were calculated at 293 K.

N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45 41

3.2.2. Time-resolved measurement in presence of surfactantsThe fluorescence decay behavior of Lum-anion was measured

with increasing concentration of surfactants in the three casesand found to follow single exponential fitting always. Some ofthe representative decay traces in different surfactant media areshown in Supplementary section (Fig. S2). The variation in fluores-cence decay time with increasing surfactant concentration in eachcase is shown in Fig. 4 and some representative data are given inTable 1. It is clear that the fluorescence decay time increases sub-stantially in TX-100 micelle (sf � 6.5 ns) in comparison with that inbuffer solution (sf � 4.6 ns). On the other hand, while the increasein sf value in SDS is moderate (sf � 5.0 ns); this quantity remainsmore or less constant (�4.6 ± 0.1 ns) over the whole concentrationrange in CTAB. These observations can be directly correlated withthe discussion made above and consistent with binding behaviorof the probe in different micelle structures. Further, the radiativeand non-radiative decay properties of Lum-anion were calculatedfrom the measured quantum yield (/f) values using the followingrelations and some of the values are given in Table 1.

jr ¼/f

sf;jnr ¼

ð1� /f Þsf

ð5Þ

The variation of total non-radiative decay parameter withincreasing concentration of surfactants is also shown in Fig. 4. Itis quite clear that the increase in fluorescence intensity in TX-100 and SDS micelles discussed earlier is direct manifestation ofthe reduction in total non-radiative relaxation pathway. The strongbinding of the probe (as evidenced by the higher KS values) inside

these micellar structures arrests several possible vibrational de-grees of freedom that results the increase in fluorescence intensity.However, in case of CTAB, both the sf and jnr values remain moreor less constant over the entire concentration range. So, thequenching of Lum-anion fluorescence with increasing concentra-tion of CTAB is definitely due to a ground state or static type ofCoulombic interaction between the oppositely charged fluoro-phore and surfactant head group towards the formation of a darkstate inside the weakly bound interfacial layer.

In this connection, it is worth mentioning a recent study re-ported by Kondo et al. on the slow dynamics of nanoconfined watermolecules in the interfacial layer of SDS and CTAB surfactants usingthe photophysical behavior of cationic dye auramine O (AuO) [43].Despite significant differences between the chemical nature aswell as photophysical behavior of AuO and Lum-anion, the trendin the variation of fluorescence intensity and/or lifetime of thesetwo probes are surprisingly similar in modulated environment ofmicellar nanocavities. These observations lead to believe more to-ward the speculation that although the excited state properties (orreactions) of probe molecules change (proceed) through a complexpotential energy surface with addition of surfactants, the co-ordi-nate involving the local water structure confined in micelle-waterinterfacial layer still plays a dominant role.

3.3. Spectral properties in presence of bovine serum albumin

The variation of Lum-anion fluorescence intensity with modelwater soluble protein, bovine serum albumin (BSA) is shown inFig. 5. Although the fluorescence peak position does not showany appreciable shift, it is found that with gradual addition ofthe protein, Lum-anion fluorescence intensity increases continu-ously and levels off after a certain concentration. In accordancewith the discussion made above in the cases of TX-100 and SDS,the increase in Lum-anion fluorescence in protein environmentcan also be considered as due to the binding of the probe proteinbinding region. Ligand replacement reaction with bilirubin (notshown) has further confirmed the binding of Lum-anion in sub-do-main IIA of the hydrophobic binding region, in consistent with ourearlier reports for interaction of similar heterocyclic systems inBSA and/or HSA [17]. BSA is well known as a transport proteinfor the anions and can be regarded as an efficient host for theLum-anion, as indeed confirmed from the calculated thermody-namic parameters (see below).

3.3.1. Thermodynamic parameters for the ligand bindingThe association constant for Lum-anion binding with BSA can

be evaluated by modified Benesi–Hildebrand (B–H) equation for1:1 interaction [44]:

1DF¼ 1

DFmaxþ 1

DFmax� 1

KP � ½BSA� ð6Þ

where DF (=F � F0) and DFmax (=Fa � F0) indicate the relative andmaximum increase in fluorescence intensity, respectively; KP isthe binding constant and [BSA] indicates the protein concentration.Rearrangement of the above equation leads to the following form

ðF1 � F0ÞðF � F0Þ

¼ 1þ 1KP� 1½BSA� ð7Þ

Therefore, a plot of (Fa � F0)/(F � F0) against 1/[BSA] would leadto a straight line with binding constant KP = (slope)�1. Linearity ofthis plot would also confirm the 1:1 stoichiometry of the probe-protein complex. Some representative plots are shown in Fig. 6and the corresponding KP values are listed in Table 3. The linearityof the plots in Fig. 6 indeed confirms the 1:1 stoichiometry of theLum-anion and BSA complex. The association constant values are

Fig. 4. Variation in Lum-anion fluorescence decay time (open squares) and total non-radiative decay rate constant (solid square) with increasing concentration of TX-100 (a),CTAB (b) and SDS (c).

Fig. 5. Variation in Lum-anion fluorescence spectral profile (kexc = 365 nm) withincreasing concentration of BSA. [BSA]/lM = 0.0 (i), 6.0 (ii), 9.0 (iii), 12.0 (iv), 15.0(v), and 24.0 (vi). Inset shows the change in emission intensity at 455 nm.

42 N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45

quite high and fall within the normal range reported earlier forsuch type of complex formation [45,46]. It is to be noted here thatthe binding constant values reported in Table 3 reflect the bindingproperties of Lum-anion in a partially unfolded BSA structure sincethe protein loses its native conformation in alkaline medium[47,48]. However, it is already known that the relative susceptibil-ity of the three ligand binding domains in BSA toward alkalinedenaturation follows the order: domain I > domain III > domain II[49]. Therefore, the KP values for Lum-anion binding (Table 3),which essentially binds in sub-domain IIA of BSA as discussed be-fore, are expected not to deviate too much from the correspondingvalues for native protein structure.

The interaction forces between drugs and bio-molecules mayinclude electrostatic interactions, formation of multiple hydrogenbonds, van der Waals interaction, hydrophobic and steric contactswithin the antibody binding site, etc. [50]. The sign and magnitudeof the thermodynamic parameters can account for the main fac-tor(s) responsible towards the stability of drug–protein complex[51]. In order to elucidate the nature of the interaction of Lum-an-ion with BSA, the thermodynamic parameters were calculatedusing van’t Hoff relation [52]. Assuming the change in enthalpy(DH) for the protein–ligand binding process to be insignificant overthe temperature range studied, the association constant (KP) andthe change in other thermodynamic parameters with temperaturecan be written by following relations:

ln KP ¼DSR� DH

RTð8Þ

DG ¼ DH � TDS ð9Þ

The enthalpy (DH) and entropy (DS) change was calculatedfrom the slope and intercept of the van’t Hoff plot (Fig. 7) at fourdifferent temperatures, viz. 298, 303, 308 and 313 K; whereas,Gibb’s free energy change (DG) can be estimated from Eq. (9). Allthe thermodynamic parameters are collected in Table 3.

The negative value of the free energy change (DG) is indicativeof a spontaneous binding of Lum-anion to BSA. Further, this exo-thermic process is accompanied by positive DS value. A positiveDS value is often indicative of a hydrophobic mechanism indrug–protein interaction [51]. Specific electrostatic interactionamong ionic species in solution is characterized by positive DSand negative DH values; whereas, negative entropy and enthalpychanges indicate the importance of van der Waals force as wellas hydrogen bond formation. Therefore, the binding of Lum-anionwith BSA predominantly involves hydrophobic interaction.

Fig. 6. Plot of (Fa � F0)/(F � F0) with different BSA concentration at 298 K (a), 303 K (b), 308 K (c) and 313 K (d). The linear regression of the experimental data with Eq. (7) isshown by the solid line along with the magnitude of slope and correlation coefficient in each case.

Table 3Binding constant (KP) values and other relative thermodynamic parameters for the interaction of Lum-anion with BSA in absence and presence of urea.a

Temp. (K) BSA

[urea] = 0 [urea] = 7 M

KP/104 (M�1) DG DH DS KP/104 (M�1) DG DH DS

298 4.98 �26.52 30.10 0.19 12.73 �29.13 �24.96 0.014303 6.99 �27.47 11.14 �29.20308 7.63 �28.42 9.56 �29.27313 9.23 �29.37 7.96 �29.34

a Gibb’s free energy (DG) and enthalpy (DH) changes are given in kJ mol�1 and entropy (DS) change is given in kJ mol�1 K�1. Values are within ±15%.

Fig. 7. van’t Hoff plot for Lum-anion binding with BSA in absence (a) and presence of 7 M urea (b). The parameter of linear regression and also the correlation coefficient isgiven in each case.

N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45 43

44 N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45

3.3.2. Effect of urea on Lum-anion binding to BSAUrea is the most widely used denaturating agent, being exten-

sively applied in biochemistry, not only to denaturate proteins athigh urea concentration (6–8 M), but also to promote controlledfolding. Whether urea interacts directly in the biopolymer promot-ing a better solvation of the apolar residues or indirectly bydecreasing the water structure is the main topic of discussion inthis research area [53,54]. Other mechanisms have also been pro-posed to explain the effect of urea in biopolymers and bioaggre-gates [55,56]. The generally accepted mechanism of urea inducedprotein denaturation involves (i) unfolding of domain III, (ii) minorstructural transformation of domain II and (iii) the separations ofsub-domains of domain III from each other [57–60]. In the presentsection, we look into the details of Lum-anion binding into partiallydenatured BSA in presence of 7 M urea.

Similar to the observation made above, the Lum-anion fluores-cence intensity increases on gradual addition of BSA in presence ofurea (Fig. S3 in Supplementary section) confirming the binding ofthe probe even in the unfolded conformation of protein. The linear-ity of B–H plot (shown in Fig. S4 of the Supplementary section) basedon Eq. (7) still confirms 1:1 binding. The binding constant values andthe corresponding thermodynamic parameters are displayed in Ta-ble 3. Interestingly, although the KP values of Lum-anion in the dena-tured protein condition are of similar order in magnitude to those infolded tertiary protein structure, the trend in variation of thesequantities with temperature in two cases are distinctly different(Fig. 7). Analyses based on Eqs. (8) and (9) lead to very small positivevalue of DS with a negative DH value (Table 3) indicating the impor-tance of electrostatic interaction in this binding process. Consideringthe anionic nature of the probe, it is reasonable to believe that itbinds with anionic binding pocket of the protein containing rela-tively excess positive charge. Although the binding constant valuesof Lum-anion with both the folded or unfolded protein are almostsimilar in magnitude, hydrophobic interaction is the principle driv-ing force for the probe binding with unfolded protein tertiary struc-ture; however, in the denatured condition, the electrostaticinteraction become prominent leading to completely different ther-modynamic behavior of binding.

3.3.3. Time-resolved fluorescence decay behaviorThe fluorescence decay times of Lum-anion with varying BSA

concentration were measured both in presence and absence of urea.Some of the decay traces are shown in Supplementary section(Fig. S2). In all the cases, single exponential decay fitting were ob-tained corresponding to the Lum-anion emission as discussed be-fore. However, interestingly, the measured sf values increase withincrease in protein concentration (from 4.6 ± 0.1 ns in buffer solu-tion to 5.1 ± 0.1 ns in presence of 24 lM BSA) in absence of urea.On the other hand, in presence of 7 M urea, the fluorescence decaytime remain more or less constant over the whole range of proteinconcentration (Fig. S5 in Supplementary section). The total non-radiative decay rate constants (jnr) were also calculated in each caseby using relations given in Eq. (5) and shown in the same figure. Themagnitude of jnr decreases regularly in absence of urea; however, itremains more or less invariant in presence of urea. Overall, theseobservations are consistent with the discussion made above in caseof surfactants and further ensures that the binding of urea in proteinsub-domain is mainly controlled by the hydrophobic interaction;whereas, in presence of urea, electrostatic force plays a major rolein the binding of Lum-anion with BSA.

4. Conclusions

The excited-state photophysical behavior of mono-anionic lum-azine (Lum-anion) has been studied by steady-state and time-re-

solved fluorescence spectroscopy in surfactants with differentlycharged head groups and also in bovine serum albumin. The fluo-rescence intensity increases in presence of non-ionic/anionic sur-factants due to the binding of the probe in hydrophobicinterfacial layer in micellar structure resulting substantial reduc-tion in non-radiative decay rate processes. However, in CTAB, theLum-anion fluorescence quenches on micelle formation, mostlydue to Coulombic interaction between the oppositely chargedprobe and micellar head groups in the ground state. Binding ofLum-anion in ligand binding domain IIA of BSA is thermodynami-cally favorable and mainly controlled by hydrophobic interactionas evidenced by positive DH and DS values. However in presenceof strong denaturing agent like urea, the binding mechanism is en-tirely different and mainly contributed through the electrostaticinteraction.

Acknowledgement

NSM thanks Council of Scientific and Industrial Research (CSIR),Govt. of India for a fellowship. Thanks are also due to Departmentof Science & Technology (DST) and University Grants Commission(UGC), Govt. of India for supporting the Chemistry Departmentthrough FIST and SAP program, respectively.

Appendix A. Supplementary material

Variation of Lum-anion fluorescence with increasing concentra-tion of TX-100 (Fig. S1), time-resolved decay traces in presence ofsurfactants/proteins (Fig. S2) and details of Lum-anion fluores-cence results in BSA denatured with 7 M urea (Figs. S3–S5). Supple-mentary data associated with this article can be found, in theonline version, at http://dx.doi.org/10.1016/j.jphotobiol.2013.02.008.

References

[1] W. Pfleiderer, Natural pteridines – a chemical hobby, in: J.E. Ayling, M.G. Nair,C.M. Baugh (Eds.), Chemistry and Biology of Pteridines and Folates, PlenumPress, New York, 1993, pp. 1–16.

[2] T.J. Kappock, J.P. Caradonna, Pterin-dependent amino acid hydroxylases, Chem.Rev. 96 (1996) 2659–2756.

[3] H. Rembold, W.L. Gyure, Biochemistry of the pteridines, Angew. Chem. Int. Ed.Engl. 11 (1972) 1061–1072.

[4] J. Lee, I.B.C. Matheson, F. Müller, D.J. O’Kane, J. Vervoort, A.J.W.G. Visser, Themechanism of bacterial bioluminescence, in: F. Müller (Ed.), Chemistry andBiochemistry of Flavoenzymes, vol. 1, CRC Press, Boca Raton, FL, 1991, pp. 109–151.

[5] B. Wörsdörfer, L.M. Henning, R. Obexer, D. Hilvert, Harnessing proteinsymmetry for enzyme design, ACS Catal. 2 (2012) 982–985.

[6] H.-N. Chen, K.J. Woycechowsky, Conversion of a dodecahedral protein capsidinto pentamers via minimal point mutations, Biochemistry 51 (2012) 4704–4712.

[7] S. Lilavivat, D. Sardar, S. Jana, G.C. Thomas, K.J. Woycechowsky, In vivoencapsulation of nucleic acids using an engineered nonviral protein capsid, J.Am. Chem. Soc. 134 (2012) 13152–13155.

[8] B. Wörsdörfer, Z. Pianowski, D. Hilvert, Efficient in vitro encapsulation ofprotein cargo by an engineered protein container, J. Am. Chem. Soc. 134 (2012)909–911.

[9] V. Zylberman, P.O. Craig, S. Klinke, B.C. Braden, A. Cauerhff, F.A. Goldbaum,High order quaternary arrangement confers increased structural stability toBrucella sp. lumazine synthase, J. Biol. Chem. 279 (2004) 8093–8101.

[10] R. Ladenstein, M. Schneider, R. Huber, H.D. Bartunik, K. Wilson, K. Schott, A.Bacher, Heavy riboflavin synthase from Bacillus subtilis. Crystal structureanalysis of the icosahedral beta 60 capsid at 3.3 A resolution, J. Mol. Biol. 203(1988) 1045–1070.

[11] L. Lu, D. Xing, N. Ren, Bioreactor performance and quantitative analysis ofmethanogenic and bacterial community dynamics in microbial electrolysiscells during large temperature fluctuations, Environ. Sci. Technol. 46 (2012)6874–6881.

[12] W.I. Pfleiderer, I. Pteridine, Über 2.4-dioxo-tetrahydropteridine, Chem. Ber. 90(1957) 2582–2587.

[13] M.P. Denofrio, A.H. Thomas, A.M. Braun, E. Oliveros, C. Lorente, Comment on‘‘Excited-state intermolecular proton transfer of lumazine’’, J. Phys. Chem. C114 (2010) 14307–14308.

N.S. Moyon et al. / Journal of Photochemistry and Photobiology B: Biology 121 (2013) 37–45 45

[14] N.S. Moyon, P.M. Gashnga, S. Phukan, S. Mitra, Specific solvent effect onlumazine photophysics: a combined fluorescence and intrinsic reactioncoordinate analysis, Chem. Phys., (2013), in press.

[15] N.S. Moyon, A.K. Chandra, S. Mitra, Effect of solvent hydrogen bonding onexcited-state properties of luminol: a combined fluorescence and DFT study, J.Phys. Chem. A. 114 (2010) 60–67.

[16] N.S. Moyon, S. Mitra, Fluorescence solvatochromism in lumichrome andexcited-state tautomerization: a combined experimental and DFT study, J.Phys. Chem. A. 115 (2011) 2456–2464.

[17] N.S. Moyon, S. Mitra, Luminol fluorescence quenching in biomimickingenvironments: sequestration of fluorophore in hydrophobic domain, J. Phys.Chem. B. 115 (2011) 10163–10172.

[18] M. Marchena, M. Gil, C. Martin, J.A. Organero, F. Sanchez, A. Douhal, Stabilityand photodynamics of lumichrome structures in water at different pHs and inchemical and biological caging media, J. Phys. Chem. B 115 (2011) 2424–2435.

[19] C. Lorente, A.H. Thomas, Photophysics and photochemistry of pterins inaqueous solution, Acc. Chem. Res. 39 (2006) 395–402.

[20] M.P. Denofrioa, S. Hatz, C. Lorente, F.M. Cabrerizo, P.R. Ogilby, A.H. Thomas,The photosensitizing activity of lumazine using 20-deoxyguanosine 50-monophosphate and HeLa cells as targets, Photochem. Photobiol. Sci. 8(2009) 1539–1549.

[21] M.P. Denofrio, A.H. Thomas, A.M. Braun, E. Oliveros, C. Lorente, Photochemicaland photophysical properties of lumazine in aqueous solutions, J. Photochem.Photobiol. A 200 (2008) 282–286.

[22] A.J.W.G. Visser, J. Lee, Lumazine protein from the bioluminescent bacteriumPhotobacterium phosphoreum. A fluorescence study of the protein-ligandequilibrium, Biochemistry 19 (1980) 4366–4372.

[23] J. Lee, Lumazine protein and the excitation mechanism in bacterialbioluminescence, Biophys. Chem. 48 (1993) 149–158.

[24] B. Valeur, in: Molecular Fluorescence Principles and Applications, Wiley-VCH,Weinheim, FRG, 2002.

[25] J.R. Lakowicz, in: Principles of Fluorescence Spectroscopy, third ed., Springer,Singapore, 2006.

[26] S.R. Meech, D. Phillips, Photophysics of some common fluorescence standards,J. Photochem. 23 (1983) 193–217.

[27] N. Mataga, T. Kubota, in: Molecular Interactions and Electronic Spectra, MarcelDekker Inc., New York, 1970.

[28] M.C. Cuquerella, M.A. Miranda, F. Bosca, Role of excited state intramolecularcharge transfer in the photophysical properties of norfloxacin and itsderivatives, J. Phys. Chem. A 110 (2006) 2607–2612.

[29] E. Maligaspe, T. Kumpulainen, H. Lemmetyinen, N.V. Tkachenko, N.K.Subbaiyan, M.E. Zandler, F. D’Souza, Ultrafast singlet�singlet energy transferin self-assembled via metal�ligand axial coordination of free-baseporphyrin�zinc phthalocyanine and free-base porphyrin�zincnaphthalocyanine dyads, J. Phys. Chem. A 114 (2010) 268–277.

[30] S. Campidelli, R. Deschenaux, A. Swartz, G.M. Aminur Rahman, D.M. Guldi, D.Milic, E. Vázqueze, M. Prato, A dendritic fullerene–porphyrin dyad, Photochem.Photobiol. Sci. 5 (2006) 1137–1141.

[31] Details of the laser strobe technique is described on the manufacture web site.<http://www.pti-nj.com>.

[32] R. Klein, I. Tatischeff, Tautomerism and fluorescence of lumazine, Photochem.Photobiol. 45 (1987) 55–65.

[33] D. Prukała, E. Sikorska, J. Koput, I. Khmelinskii, J. Karolczak, M. Gierszewski, M.Sikorski, Acid–base equilibriums of lumichrome and its 1-methyl, 3-methyl,and 1,3-dimethyl derivatives, J. Phys. Chem. A 116 (2012) 7474–7490.

[34] I. Presiado, Y. Erez, R. Gepshtein, D. Huppert, Excited-state intermolecularproton transfer of lumazine, J. Phys. Chem. C 114 (2010) 3634–3640.

[35] R.S. Sarpal, M. Belletête, G. Durocher, Fluorescence probing and proton-transfer equilibrium reactions in water, SDS, and CTAB using 3,3-dimethyl-2-phenyl-3H-indole, J. Phys. Chem. 97 (1993) 5007–5013.

[36] C. Hirose, L. Sepulveda, Transfer free energies of p-alkyl-substituted benzenederivatives, benzene, and toluene from water to cationic and anionic micellesand to n-heptane, J. Phys. Chem. 85 (1981) 3689–3694.

[37] T.S. Singh, S. Mitra, Fluorimetric studies on the binding of 4-(dimethylamino)cinnamic acid with micelles and bovine serum albumin,Photochem. Photobiol. Sci. 7 (2008) 1063–1070.

[38] S. Phukan, S. Mitra, Fluorescence behavior of ethidium bromide inhomogeneous solvents and in presence of bile acid hosts, J. Photochem.Photobiol. A: Chem. 244 (2012) 9–17.

[39] A. Mallick, P. Purkayastha, N. Chattopadhyay, J. Photochem. Photobiol. C: Rev.8 (2007) 109–127.

[40] H.H. Paradies, Shape and size of a nonionic surfactant micelle. Triton X-100 inaqueous solution, J. Phys. Chem. 84 (1980) 599–607.

[41] S.S. Berr, Solvent isotope effects on alkytrimethylammonium bromide micellesas a function of alkyl chain length, J. Phys. Chem. 91 (1987) 4760–4765.

[42] M. Appell, W.B. Bosma, Effect of surfactants on the spectrofluorimetricproperties of zearalenone, J. Lumin. 131 (2011) 2330–2334.

[43] M. Kondo, I.A. Heisler, S.R. Meech, Reactive dynamics in micelles: auramine Oin solution and adsorbed on regular micelles, J. Phys. Chem. B 114 (2010)12859–12865.

[44] H.A. Benesi, J.H. Hildebrand, Spectrophotometric investigation of theinteraction of iodine with aromatic hydrocarbons, J. Am. Chem. Soc. 71(1949) 2703–2707.

[45] S. Ghosh, S. Jana, N. Guchhait, Domain specific association of small fluorescentprobe trans-3-(4-monomethylaminophenyl)-acrylonitrile (MMAPA) withbovine serum albumin (BSA) and its dissociation from protein binding sitesby Ag nanoparticles: spectroscopic and molecular docking study, J. Phys.Chem. B 116 (2012) 1155–1163.

[46] A. Mallick, B. Halder, N. Chattopadhyay, Spectroscopic investigation on theinteraction of ICT probe 3-acetyl-4-oxo-6,7-dihydro-12H indolo-[2,3-a]quinolozine with serum albumins, J. Phys. Chem. B 109 (2005) 14683–14690.

[47] M.C. Hilak, B.J.M. Harmsen, W.G.M. Braam, J.J.M. Joordens, G.A.J. Van Os,Conformational studies on large fragments of bovine serum albumin inrelation to the structure of the molecule, Int. J. Peptide Protein Res. 6 (1974)95–101.

[48] F. Shahid, J.E. Gomez, E.R. Birnbaum, D.W. Darnall, The lanthanide-induced N =to F transition and acid expansion of serum albumin, J. Biol. Chem. 257 (1982)5618–5622.

[49] B. Ahmad, M.Z. Kamal, R.H. Khan, Alkali-induced conformational transition indifferent domains of bovine serum albumin, Protein Pept. Lett. 11 (2004) 307–315.

[50] D. Leckband, Measuring the forces that control protein interactions, Ann. Rev.Biophys. Biomol. Struct. 29 (2000) 1–26.

[51] P.D. Ross, S. Subramanian, Thermodynamics of protein association reactions:forces contributing to stability, Biochemistry 20 (1981) 3096–3102.

[52] R.J. Silbey, R.A. Alberty, in: Physical Chemistry, third ed., John Wiley & Sons(Asia) Pte. Ltd., Singapore, 2002.

[53] O. Enea, C. Jollcoeur, Heat capacities and volumes of several oligopeptides inurea-water mixtures at 25 �C. Some implications for protein unfolding, J. Phys.Chem. 86 (1982) 3870–3881.

[54] E. Liepinsh, G. Otting, Specificity of urea binding to proteins, J. Am. Chem. Soc.116 (1994) 9670–9674.

[55] G.D. Dias, F.H. Florenzano, W.F. Reed, M.S. Baptista, S.M.B. Souza, E.B. Alvarez,H. Chaimovich, I.M. Cuccovia, C.L.C. Amaral, C.R. Brasil, L.S. Romsted, M.J.Politi, Effect of urea on biomimetic systems: neither water 3-D structurerupture nor direct mechanism, simply a more ‘‘polar water’’, Langmuir 18(2002) 319–324.

[56] J. Gradolnik, Y. Maréchal, Urea and urea–water solutions—an infrared study, J.Mol. Struct. 615 (2002) 177–189.

[57] J. González-Jiménez, M. Cortijo, Urea-induced denaturation of human serumalbumin labeled with acrylodan, J. Protein Chem. 21 (2002) 75–79.

[58] A. Sułkowska, B. Bojko, J. Rownicka, D. Pentak, W. Sulkowski, Effect of urea onserum albumin complex with antithyroid drugs: fluorescence study, J. Mol.Struct. 651 (653) (2003) 237–243.

[59] M.Y. Khan, S.K. Agarwal, S. Hangloo, Urea-induced structural transformationsin bovine serum albumin, J. Biochem. 102 (1987) 313–317.

[60] B. Ahmad, M.K.A. Khan, S.K. Haq, R.H. Khan, Intermediate formation at lowerurea concentration in ‘B’ isomer of human serum albumin: a case study usingdomain specific ligands, Biochem. Biophys. Res. Commun. 314 (2004) 166–173.