Lysozyme photo-oxidation by singlet oxygen: properties of the partially inactivated enzyme

1011-1344/00/$ - see front matter q2000 Elsevier Science S.A. All rights reserved.PII S1011- 1344 (00)00049 -X

Tuesday Jun 13 10:07 AM StyleTag -- Journal: JPB (J. Photochem. Photobiol. B: Biol.) Article: 7958

www.elsevier.nl/locate/jphotobiol

J. Photochem. Photobiol. B: Biol. 55 (2000) 196–200

Lysozyme photo-oxidation by singlet oxygen: properties of the partiallyinactivated enzyme

Eduardo Silva a,*, Constanza De Landea a, Ana Marıa Edwards a, Eduardo Lissi b´a Departamento de Quımica Fısica, Facultad de Quımica, Pontificia Universidad Catolica de Chile, Casilla 306, Correo 22, Santiago, Chile´ ´ ´ ´

b Departamento de Quımica, Facultad de Quımica y Biologıa, Universidad de Santiago de Chile, Casilla 40, Correo 33, Santiago, Chile´ ´ ´

Received 3 November 1999; accepted 9 April 2000

Abstract

This work studies the behaviour of partially inactivated lysozyme formed by the effect of singlet oxygen, which was obtained through theirradiation of the native enzyme solution with polychromatic visible light using Methylene Blue as a sensitizer. The polyacrylamide gelanalysis of the partially inactivated lysozyme solution shows the presence of different protein fractions. One of them, which corresponds to53% of the original enzyme, has the same migration as the native enzyme. The others are a mixture of fractions (47%) that show slowermigration to the cathode. When this experiment is carried out in the presence of sodium dodecyl sulfate, only one fraction is obtained, whichrules out the presence of covalently aggregated forms of lysozyme. The partially inactivated lysozyme has lost 74% of the fluorescenceemission of the tryptophan (Trp) residues. By using the anionic quencher iodide, it is determined that 45 and 36% of the fluorescence emissionarising from the native and partially inactivated enzyme, respectively, are due to Trp residues exposed to the solvent. Michaelis–Mentenconstants (Km) of 0.296 and 0.511 (mg/ml) and maximum initial rates (Vmax) of 0.295 and 0.190 (mg/ml min) are determined for thenative and the partially inactivated enzyme solutions, respectively. The same inactivation profile is found when the denaturing effect ofincreasing urea concentration on both the native and partially inactivated lysozyme is studied. It is postulated that the partially inactivatedlysozyme solution is composed of one protein fraction with enzymatic activity similar to that of the native enzyme and also of a mixture offractions (47% of the total enzyme) with very low activity and characterized by a high tryptophan photo-oxidation. q2000 ElsevierScience S.A. All rights reserved.

Keywords: Lysozyme; Enzyme inactivation; Singlet oxygen; Methylene Blue; Michaelis–Menten kinetics

1. Introduction

Under normal physiological conditions, the use of oxygenby cells of aerobic organisms generates potentially deleteri-ous reactive oxygen metabolites [1].

Species such as hydroxyl radicals, peroxide radicals, anionsuperoxide, singlet oxygen and ozone are important media-tors of cell injuries. Initially, cell damage was fundamentallyattributed to peroxidative processes experienced by cellmem-branes. Nevertheless, recently it has been shown that theoxidative damage of DNA [2–4] and proteins [5,6] fre-quently precedes and is more difficult to prevent than thelipoperoxidative process.

One way that molecular oxygen (3O2) can becomereactiveis by flipping the spin of one of the two unpaired electrons tobecome singlet oxygen (1O2). Conversion of 3O2 to 1O2

requires at least 22.7 kcal/mol of energy, corresponding to a

* Corresponding author. Tel. q56-2-686-4394; fax: q56-2-686-4744;e-mail: [email protected]

near-infrared photon (1270 nm) or to three times the energyreleased in the hydrolysis of ATP to ADP. This energy cancome from a variety of chemical or photochemical reactions,in vivo as well as in vitro, one of the most common beingenergy transfer to oxygen from another excited molecule. Incontrast to 3O2,

1O2 reacts rapidly and selectively with high-electron-density sites in many unsaturated organic com-pounds (e.g., some free or protein-bound amino acids) viaspin-allowed processes resulting in various peroxides(hydroperoxides and endoperoxides); these can react further,often yielding free radicals [7,8]. In biological systems, thereare two major sources of 1O2: the inflammatory process andphotosensitization in light-exposed areas [8]. The reactivityof singlet oxygen against proteins [9,10], peptides and aminoacids [11–14] has recently been studied.

The dye-sensitized photo-oxidation of proteins has beenwidely used for the modification of specific amino-acid res-idues in order to elucidate their role in biological functioning[15,16]. The possibility of correlating the data of dye-sen-

E. Silva et al. / J. Photochem. Photobiol. B: Biol. 55 (2000) 196–200 197


sitized photo-oxidation with the three-dimensional confor-mation of proteins was first suggested by Ray and Koshland[17]. In all this work it was assumed that the solution of theinactivated enzyme consisted of a homogeneous populationof the modified protein. In this paper we aim to study if the1O2-mediated change in activity of a model enzyme, lyso-zyme, is an ‘all or nothing’ process, or if the loss of activityof a given enzyme is a progressive process in which partiallymodified molecules show intermediate activities. In particu-lar, we analyse the catalytic behaviour (Michaelis–Mentenparameters), the denaturation by urea and the exposure to thesolvent of the tryptophan (Trp) residues of native and par-tially inactivated solutions of lysozyme.

2. Materials and methods

2.1. Materials

Deuterium oxide, salt-free egg-white lysozyme, Methyl-ene Blue and Micrococcus lysodeikticus were obtained fromSigma (St. Louis, MO). Sephadex G-25 was obtained fromPharmacia (Uppsala, Sweden). The micro bicinchoninicacidprotein assay reagent kit was purchased from Pierce (Rock-ford, IL). All other reagents were of analytical grade.

2.2. Irradiation conditions

Solutions of proteins (1.0 mg/ml) and Methylene Blue(3.5=10y5 M), in 0.05 M phosphate buffer, pH 7.0, wereirradiated, under oxygen bubbling, with polychromatic visi-ble light from a 150 W slide projector lamp (Osram, Xeno-phot HLX) in a 1 cm light-path cell, perspex water jacketedat 378C.

2.3. Enzymatic activity assay

The activity of the enzyme was assayed by measuring thelysis of Micrococcus lysodeikticus [18].

Micrococcus lysodeikticus, 0.2 mg/ml, was suspended in3 ml of 0.07 M phosphate buffer, pH 7.0, which was 0.017M in NaCl. The reaction was started by adding 10 ml of theenzyme solution. The decrease in optical density at 436 nmwas spectrophotometrically recorded. In order to determinethe enzymatic rate at different substrate concentrations, it wasnecessary to demonstrate that there is a direct relationbetween the turbidity of the solution at 436 nm and the Micro-coccus lysodeikticus concentration. This fact was experimen-tally observed for the range of substrate concentrationsemployed in this work (0.05–0.20 mg/ml).

2.4. Analysis by sodium dodecyl sulfate polyacrylamide gelelectrophoresis (SDS–PAGE)

SDS–PAGE (7.5% acrylamide and 0.2% bis acrylamide)was conducted according to the procedure described by

Laemmli [19] in 1 mm thick gels in an LKB 2117 Multiphorgel apparatus (Uppsala, Sweden), using 0.2 M phosphatebuffer, pH 7.1. Protein fractions were dissolved in 0.2 Mphosphate buffer, pH 7.1, containing 1% v/v 2-mercapto-ethanol and 1% w/v sodium dodecyl sulfate (SDS). Thedissolved sample was incubated at 378C for 8 h.

Polyacrylamide gels in the absence of SDS were preparedusing 10% acrylamide and 0.2% bis acrylamide in Glycine–Temed buffer, pH 7.3.

The densitometric analysis of the gels was carried out usingan Elphor densitometer (Munich, Germany).

2.5. Gel filtration

The protein was separated from the sensitizer through aSephadex G-25 column (1 cm=5 cm), which had beenequilibrated with 0.05 M phosphate buffer, pH 7.0.

2.6. Protein assay

The protein fractions eluted from the Sephadex G-15 col-umn were adjusted to the same protein concentration,employing the bicinchoninic acid formulation for the quan-titative colorimetric determination of total protein in diluteaqueous solutions [20].

2.7. Fluorescence quenching experiments

Quenching measurements were carried out for eight sam-ples of the native enzyme and 50% inactivated lysozyme,with varying amounts of sodium iodide (0–0.8 M). Theywere prepared by diluting stock solutions of the model com-pound, of NaI and of NaCl, all prepared in 0.05 M phosphatebuffer, pH 7.0. NaCl was used to maintain the ionic strength.All solutions were freshly prepared, and a small amount ofS2O3

2y was added to the iodide stock solution to prevent I3y

formation. The solutions were equilibrated at 258C beforeand during the measurements. Fluorescence intensities weremeasured in a Perkin–Elmer 650-10S fluorescence spectro-photometer. An excitation wavelength of 295 nm was usedin all cases to ensure selective excitation of the Trp residues.

3. Results

Irradiation of a lysozyme solution with visible light at378C, in the presence of Methylene Blue, leads to a progres-sive loss of enzymatic activity (Fig. 1). This inactivation issignificantly enhanced when the H2O of the buffer solutionis replaced by D2O. From the kinetic data shown in Fig. 1, itwas possible to select an irradiation time corresponding to50% enzyme inactivation (activity measured at 0.2 mg/mlsubstrate). The native enzyme and the 50% inactivated lyso-zyme were separated from the sensitizer through a molecularexclusion column. Both solutions were adjusted to the sameprotein concentration [20] and their absorbances determined

E. Silva et al. / J. Photochem. Photobiol. B: Biol. 55 (2000) 196–200198


Fig. 1. Inactivation of lysozyme (7.0=10y5 M) upon visible-light photo-oxidation, sensitized by Methylene Blue (3.5=10y5 M). The enzyme solu-tions were irradiated in aqueous or D2O 0.05 M phosphate buffer, pH 7.0.

Fig. 2. Stern–Volmer plots for iodide quenching of tryptophan fluorescencefrom native and partially inactivated lysozyme.

Fig. 3. Modified Stern–Volmer plots for iodide quenching of tryptophanfluorescence from native and partially inactivated lysozyme.

Fig. 5. Lineweaver–Burk plots for native and partially inactivated lysozyme.

Fig. 4. Michaelis–Menten plots for native and partially inactivated lysozyme.

at 286 nm. It was found that the 50% inactivated enzyme hadonly 74% of the absorbance at 286 nm compared with that ofthe native enzyme. The fluorescence emission intensity(lexcs295 nm) of the inactivated enzyme was only 19% ofthat of the native enzyme.

In order to study the correlation of the accessibility of theTrp residues of lysozyme with the fluorescence intensity,afterthe Methylene Blue photosensitized oxidation, fluorescencequenching studies were carried out by selectively exciting

the Trp residues at 295 nm. An ionic quencher that actspreferentially on accessible residues was employed. Theiodide ion was chosen because it is an efficient dynamicquencher [21]. Fig. 2 shows the effect of the addition ofiodide on the fluorescence intensity of native and previously50% inactivated lysozyme. The fact that a straight line in theclassical Stern–Volmer representation was not obtained istypical of Trp residues with different degrees of accessibilityto the quencher [21].

In order to determine the fraction of the total fluorescencethat results from the Trp residues exposed to the solvent, theLehrer equation [22] was used:

I /(I yI)s1/F q1/(F K [Q])0 0 w w sv

Lehrer plots are shown in Fig. 3. The intercept correspondsto 1/Fw, where Fw represents the fraction of the total fluores-cence resulting from the exposed Trp residues. In the case ofnative and previously 50% inactivated lysozyme, 45 and 36%of the fluorescence is due to Trp residues exposed to thesolvent, respectively. The Ksv values in both cases were verysimilar, 2.45 (native) and 2.48 (50% inactivated lysozyme).

Michaelis–Menten and Lineweaver–Burk plots for thenative and the 50% inactivated enzyme were performedunderthe standard conditions, using Micrococcus lysodeicticus cellwall as substrate. Fig. 4 shows that both forms of the enzymefollow a hyperbolic Michaelis–Menten relationship.Thedou-ble-reciprocal plots shown in Fig. 5 allowed Vmax and Km to

E. Silva et al. / J. Photochem. Photobiol. B: Biol. 55 (2000) 196–200 199


Table 1Maximum initial rates (Vmax) and Michaelis–Menten constants (Km) ofnative and 50% inactivated lysozyme solutions

Native lysozyme 50% inactivated lysozyme

Vmax (mg/ml min) 0.295"0.029 0.190"0.028Km (mg/ml) 0.296"0.051 0.511"0.078

Fig. 6. Polyacrylamide gel electrophoresis of a lysozyme solution irradiatedwith visible light for 0 (1), 15 (2) and 20 (3) min, respectively, in thepresence of Methylene Blue.

be determined. The values obtained through repeated exper-iments are shown in Table 1.

The results of the polyacrylamide gel electrophoresis ofnative and 50% inactivated lysozyme samples are shown inFig. 6. As determined by densitometric analysis of the gels,the irradiated lysozyme solution consisted of 53% of a frac-tion with the same migration properties as the native enzyme,and a mixture of fractions (47% of the total) with a lowermigration. Denaturation processes with an increasing con-centration of urea on the native and 50% active enzyme aresimilar, since the same inactivation profile was observed (notshown).

4. Discussion

The enzyme lysozyme was efficiently inactivated by sin-glet oxygen generated by the action of visible light, usingMethylene Blue as sensitizer. The role of this oxygen reactivespecies in the enzyme inactivation was indirectly demon-strated by employing the useful diagnostic test for singletoxygen based on the large effect of deuterium on the half-lifeof 1O2. The half-life of 1O2 is 53–68 ms in D2O as comparedwith a half-life of f4 ms in H2O [23]. Under the reactionconditions used in this work, a reaction rate increase by afactor of five was observed in going from H2O to D2O. Thisvalue is lower than the theoretical one, but represents clearevidence for the presence of singlet oxygen during thereaction.

A 50% inactivated lysozyme solution was obtained byselecting the irradiation time from the data shown in Fig. 1.The kinetic parameters of the 50% inactivated lysozyme weredifferent from those of the native enzyme. The Km value ofthe 50% inactivated lysozyme was 1.7 times greater than thatof the native enzyme, indicating a loss in the enzyme–sub-strate complex affinity. The maximum velocity Vmax of the50% inactivated solution corresponds to 64% of the value forthe native enzyme. It must be noted that 50% inactivationwas achieved using a substrate concentration of 0.2 mg/ml.At this concentration, the enzyme was not in the maximum-velocity condition.

The initial velocity (V0) of an enzyme-catalysed reactionis given by

V sk [ES] (1)0 2

and the maximum velocity occurs when the enzyme is satu-rated and [ES]s[Et] (Et is the total enzymeconcentration).Vmax can be defined as

V sk [Et] (2)max 2

The result of the polyacrylamide gel analysis of the par-tially inactivated enzyme indicates that this solution was amixture of protein fractions. One of these, which correspondsto 53% of the total enzyme, had identical electrophoreticproperties to those of the native enzyme. It must be noted thatthis fraction (53% of the original native enzyme) cannot

account for the 64% Vmax value, so the other fractions (47%of total protein) must be contributing 11% of the Vmax value.It is expected that the contribution of these low-activity frac-tions to the urea denaturation profile would be even lowerbecause these experiments were not carried out in the maxi-mum-velocity condition. The presence of different activepopulations of the enzyme can explain the poor linear cor-relation found in the Lineweaver–Burk representation of thepartially inactivated enzyme solution, due to a small down-ward curvature observed at higher substrate concentrations,which is expected for non-homogeneous populations ofactive enzymes.

Lysozyme presents six Trp and three tyrosine (Tyr) resi-dues. Considering that the extinction coefficient of Trp is 4.7times greater than that of Tyr and that Trp residues are morereactive toward singlet oxygen than Tyr residues in proteins[9,10], peptides and amino acids [11–14], we assume thatthe 26% decrease in the absorption at 286 nm of the partiallyinactivated lysozyme is mostly due to Trp photo-oxidation.If this damage is assigned exclusively to the new fraction thatappears after irradiation, it can be assumed thatapproximatelythree Trp residues are modified per enzyme molecule. On theother hand, the fact that the fluorescence intensity of thepartially inactivated lysozyme solution was only 19% of that

E. Silva et al. / J. Photochem. Photobiol. B: Biol. 55 (2000) 196–200200


of the native enzyme indicates that there is no direct corre-lation between the fluorescence decrease and enzyme deac-tivation. If the Trp residues of the higher activity fractionwere unaltered, at least 53% of the original fluorescencewould be expected. This shows that all the lysozyme fractionshave experienced some modification at the level of Trp resi-dues. It is noteworthy that the fluorescence quantum yield ofthe Trp residues in proteins varies according to their micro-environment [24]; therefore not all Trp residues contributeto the same extent to the fluorescence emission. The highfluorescence decrease must be interpreted through both Trpphoto-oxidation and alteration in the Trp environment, thelatter being produced by alterations in the protein confor-mation. From the iodide quenching studies, it can be con-cluded that the most exposed Trp residues of the nativeenzyme are the most extensively modified.

Acknowledgements

This work was supported by FONDECYT, grant no.1970691.

References

[1] R.S. Sohal, R. Weindruch, Oxidative stress, caloric restriction, andaging, Science 273 (1996) 59–63.

[2] H.D. Osieswacz, Genetic regulation of aging, J. Mol. Med. 75 (1997)715–727.

[3] B.N. Ames, R.L. Saul, E. Schwiers, R. Adelman, R. Catchcart, Oxi-dative DNA damage as related to cancer and aging: assay of thymineglycol, thymidine glycol and hydroxymethyl–uracil in human and raturine, in: R.S. Sohal, R.G. Cutler (Eds.), Molecular Biology of Aging,Raven, New York, 1985, pp. 137–144.

[4] C. Richter, J.W Park, B.N. Ames, Normal oxidative damage to mito-chondrial and nuclear DNA is extensive, Proc. Natl. Acad. Sci. USA85 (1988) 6465–6467.

[5] E.R. Stadman, Protein modification in aging, J. Gerontol. 44 (1988)B112–120.

[6] E.R. Stadman, Protein oxidation and aging, Science 257 (1992)1220–1224.

[7] A.U. Khan, T. Wilson, Reactive oxygen species as cellular messen-gers, Chem. Biol. 2 (1995) 437–445.

[8] K. Briviba, L.-O. Klotz, H. Sies, Toxic and signaling effects of pho-tochemically or chemically generated singlet oxygen in biologicalsystems, Biol. Chem. 37 (1997) 1259–1265.

[9] J.R. Kanofsky, Quenching of singlet oxygen by human plasma, Pho-tochem. Photobiol. 51 (1990) 299–303.

[10] A. Michaeli, J. Feitelson, Reactivity of singlet oxygen toward proteins:the effect of structure in basic pancreatic trypsin inhibitor and inribonuclease A, Photochem. Photobiol. 65 (1997) 309–315.

[11] S.G. Bertolotti, N.A. Garcia, G.A. Arguello, Effect of the peptide¨bond on the singlet molecular oxygen-mediated sensitized photo-oxi-dation of tyrosine and tryptophan dipeptides. A kinetic study, J. Pho-tochem. Photobiol. B: Biol. 10 (1991) 57–70.

[12] S. Miskoski, N.A. Garcia, Influence of the peptide bond on the singletmolecular oxygen-mediated (O2[

1Dg]) photooxidation of histidineand methionine dipeptides. A kinetic study, Photochem. Photobiol.57 (1993) 447–452.

[13] A. Michaeli, J. Feitelson, Reactivity of singlet oxygen toward aminoacids and peptides, Photochem. Photobiol. 59 (1994) 284–289.

[14] A. Michaeli, J. Feitelson, Reactivity of singlet oxygen toward largepeptides, Photochem. Photobiol. 61 (1995) 255–260.

[15] J. D. Spikes, R. Straight, Sensitized photochemical processes in bio-logical systems, Annu. Rev. Phys. Chem. 18 (1967) 409–436.

[16] J.D. Spikes, M. MacKnight, Dye-sensitized photo-oxidation of pro-teins, Ann. N.Y. Acad. Sci. 171 (1970) 149–162.

[17] W.J. Ray Jr., D.E. Koshland, A method for characterizing the typeand numbers of groups involved in enzyme action, J. Biol. Chem. 236(1961) 1973–1979.

[18] D. Shugar, The measurement of lysozyme activity and the ultravioletinactivation of lysozyme, Biochim. Biophys. Acta 8 (1952) 302–310.

[19] U.K. Laemmli, Cleavage of structural proteins during the assemblyof the head of bacteriophage T4, Nature 227 (1970) 680–685.

[20] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner,M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson, D.C.Klenk, Measurement of protein using bicinchoninic acid, Anal.Biochem. 150 (1985) 76–85.

[21] M.R. Eftink, C.A. Ghiron, Fluorescence quenching studies with pro-teins, Anal. Biochem. 114 (1981) 199–227.

[22] S.A. Lehrer, Solute perturbation of protein fluorescence. The quench-ing of the tryptophyl fluorescence of model compounds of lysozymeby iodide ion, Biochemistry 17 (1971) 3254–3263.

[23] C.S. Foote, E.L. Clennan, Properties and reactions of singlet oxygen,in: C.S. Foote, J.S. Valentine, A. Greenberg, J.F. Liebman (Eds.),Active Oxygen in Chemistry, Blackie Academic and Professional,London, 1995, pp. 105–141.

[24] E.A. Burstein, N.S. Vedenkina, M.N. Ivkova, Fluorescence and thelocation of tryptophan residues in protein molecules, Photochem. Pho-tobiol. 18 (1973) 263–279.

Documents

Lysozyme photo-oxidation by singlet oxygen: properties of the partially inactivated enzyme