Modification of horseradish peroxidase induced by hydroxyl radicals. The influence of oxygen

Biochimie (1996) 78, 62-65 © Soci6t6 franqaise de biochimie et biologie mol6culaire / Elsevier, Paris

Short communication

Modification of horseradish peroxidase induced by hydroxy| radicals. The .nflu_n,.,, ef oxygen L Gebicka, JL Gebicki Institute of Applied Radiation Chemistry, Technical University of L6d~, 93-590 L6d~., Wr6blewskiego 15, Poland (Received 1 June 1995; accepted 21 September 1995)

Summary m Reactions of hydroxyl radicals with horseradish peroxidase (HRP) have been studied by means of pulse radiolysis technique in the absence and presence of oxygen. Hydroxyl radicals, produced in excess towards enzyme, react exclusively with the protein part of HRP with the rate constant k = 1.1 x l0 I! M -l s -l. Activity loss induced by OH" is connected with such an enzyme modification that causes both the interference with substrate binding and partial blocking of the channel used by peroxide. It is shown that in the presence of oxygen the loss of activity is ca 10% higher, mainly due to restrictions in the formation of compound I, ie ferryl [Fe(IV)=O] 7t-radical cation. horseradish peroxidase / hydroxyl radicals / oxygen effect / pulse radiolysis

Introduction Hydroxyl radicals (OH') formed in biological systems are highly deleterious for endogenous molecules. They can be produced directly through irradiation or indirectly via the iron-driven Fenton reaction. Protein sites of known reactivity towards hydroxyl radicals are aromatic and heterocyclic peptide residues (addition) and ubiquitous CH groups (H-atom abstractitm) The resulting radicals, with few exceptions, react with oxygen to give the corresponding peroxyl radicals. It has been shown that OH'-modified proteins are easily aggregated whereas OH" + 02 caused mainly protein fragmentation [ 1]. Peroxidases (Per) are the class of enzymes utilizing hydrogen peroxide and other organic peroxides for the oxidation of a variety of substrates: Per + H202 --¢ compound I + H20 Compound I + AH2 --~ compound II + AH" Compound Ii + AH2 "-->P,~r + AH"

(1) (2) (3)

where compound I is ferryl [Fe(IV)=O] n-radical cation, comFound II is ferryl derivative of enzyme and AH2 is the oxidized molecule.

Abbreviations: HRP, horseradish peroxidase, ABTS, 2,2'azino-bis[3-ethyl-benzothiazoline.(6).sulphonic acidl.

Horseradish peroxidase (HRP) is a hemoprotein having ferriprotoporphyfin IX as a prosthetic group and imidazole in the fifth coordination position. We have shown that: i) hemoprotein catalase exhibits some enzyme activity even after complete disappearance of the native structure due to OH" action [2]; and ii) OH" radicals do not react directly with the home centrum of catalase, HRP and lactoperoxidase [3, 41. Moreover, we have observed partial reduction (10-15%) of home iron, when peroxidase was in great excess towards OH" radicals [4]. In this paper the modification of HRP caused by the excess of homogeneously produced hydroxyl radicals and the role of oxygen in enhancing OH'-induced changes are examined. Material and methods HRP (type V I ) w i t h RZ (A4JA2s0) of 3.0 was obtained from Sigma.The concentration of the enzyme was 8 ttM as determined spectrophotometrically at 403 nm using an extinction coefficient of 1.02 x 10s M-! cm-i [5]. Water from Millipore was used throughout. OH" radicals were produced radiolytically in N20-saturated aqueous solution of the enzyme, where all e-aq were converted to OH'. Combined action of OH'/O2 was observed in N20/O 2 (4:1 v/v) saturated solution. Pulse radiolysis experiments were performed with the LINAC at the Institute of Applied Radiation Chemistry, Technical University of Ldd~ (Poland). The accelerator, optics and detection system have been described elsewhere [6, 7]. Pulses

63 of 17 ns delivering a dose of 40 Gy (giving ca 22 gM OH" [81) were applied. Samples for time-resolved spectral measurements were placed in a Suprasil flow cell connected to a flow system. Inactivation studies were carried out in Suprasil rectangular cells. The stable absorption spectra of modified enzyme were recorded on Hewlett-Packard 8452 diode-a~Tay spectrophotometer. Stopped-flow spectrc,fluorim¢~er DX-I 7MV (Applied Photophysics) was used for the observation of the reaction of modified HRP with hydrogen peroxide. Enzyme activity was assayed using ABTS [9]. All the measurements have been made at least three times on three independently prepared samples. The estimated accuracy of determinations was + 10% for pulse radiolysis, + 5% for activity and stopped-flow, and _+ 2% for stable absorption spectra measurements.

ResuRs and discussion Time-resolved spectral studies All the time-resolved spectra have been recorded on the samples irradiated with a single electron pulse, ie under conditions where no activity loss is observed (see below). In competition with SCN- at 5 x 10-4 M, the rate constant for reaction of OH" with HRP is found to be 1.1 x 10 nn M -j s -~ at pH 6.5, taking k(OH" + S E N - = 1.1 x 10m M -! s q [8]. The transient spectrum with absorption in UV region and maximum around 440 nm, formed by reaction of OH" radicals with H R P is shown in figure I.

0.05-i

: °

/ °'a./

/e i

"~ -0.05.

<"~

~e i

xO.5

@ ~ , , T -

~00

|

400

i

500

600

700

Wavelength, nm

Fig 1. Absorption spectra taken: a, -80 gs and e, -40 ms after 17 ns pulse to N20-saturated 8 gM HRP solution (pH 6.5, 5 mM phosphate buffer, ambient temperature).

The band around 440 nm is relatively stable and is still observed on a millisecond time-scale, when catalytic processes take place. However, h~ the presence of o×ygen tile band disappears in microsecond time-scale (data not shown). Absorption in the UV region is connected with primary products of OH" attack on protein molecule. The m a x i m u m around 440 nm is less obvious. It is known that sulfur-centered free radicals, including three-electron bonded sulphur-other heteroatom intermediates absorb in the visible range (400--650 : -~ [10]. However, except disulphide radical anioh (RSSR-), other sulfur-centered radicals are short lived species, decaying in microseconds. RSSR-" radical anion (absorbing in the 3 8 0 - 4 5 0 nm range) can be formed via association of RS" with RS- or via oneelectron reduction of the disulfide by such a strong reductant as hydrated electron (E,, [RSSR/RSSR- = -1.6 V [11] and Eo [H20/e-aq] = -2.< V [8]) Although in the HRP molecule four disulphic, e bridges are present, there are no thiolate residues. Reduction of disulfide is impossible under our experimental conditions where all e-,q are scavenged by N~O. Thus it seems reasonable that subtle absorption changes around 440 nm are connected with the heme chromophore. As no absorption changes are observed at the Soret band (403 nm), characteristic for ~-rt* transitions in the porphyrin ring, the possibility that OH" can react directly with the porphyrin ring is ruled out. If OH" reacted with substituents on the porphyrin ring to give slight absorption changes at 440 nm, peroxy radicals or products from their conversion also should influence the absorption spectrum in this region, which is not the case. tlence, we suggest that modification of the apoprotein in a manner that slightly influences heine absorption might be responsible ~or higher absorption in the 440 nm region. The formation of compound I (reaction l) was obselwed on a millisecond time-scale. The concentration of this compound in N:O-saturated solution was about 5 g M , taking A~ = 3000 M-' cm -~ at 660 nm [12]. This relatively low conversion (ca 60%) is limited by the amount of H20,, produced in spur reactions and from OH" recombination [8]. Indeed, in the N20/O2 saturated solution we could detect 8 g M of compound I, which means 100% conversion of the enzyme. We can explain this in terms of additional H,O: generation from HOJO,-" dismutation [ 13-15]. Compound I is transformed to compound II in seconds after the pulse, most likely by intramolecular electron transfer from modified protein moiety. Absorption spectra taken 5 rain after pulse irradiation of both deoxygenated and oxygenated solutions represent a mixture of compound I[ and native enzyme. After 30 rain, compound II fully disappears and native enzyme is recovered.

64

Table L Hydroxyl radical induced HRP modifications in the ~bsence/presence of oxygen. IOH'I

RZ

(W~¢) 0 22 43 108 216 324

3.05/3.05 2.90/2.85 2.68/2.70 2.41/2.33 2.07/1.84 1.68/1.54

Relative absorption changes at 403 nm (%)

Activity loss

0/0 0/0 6/6 8/11 18/24 22/37

0/0 0/0 23/28 34/40 53/64 69/'71

Inactivation of HRP

All the inactivation measurements were made 24 h after irradiation of the samples. When HRP solution is single pulsed, no activity loss is observed (see above). However, the dose of 80 Gy (two pulses into the same sample) causes some inactivation. Table I shows the decrease of enzyme activity with the dose. OH'-induced HRP inactivation observed in this study is much more effective than that observed earlier by us after T-radiolysis of HRP at the same dose [16]. One must remember, however, that the rate of OH" formation in the pulse radiolysis is much higher than in steady-state radiolysis, where some modified enzyme molecules may have enough time to regain their active form during irradiation. The action of OH" alone or in combination with oxygen decreases the Soret band (403 nm) without a detectable shift in wavelength and increases the UV absorption (280 nm). In consequence, RZ values (A4o3/A2~o) decrease with increasing [OH']. The increase of absorbance in the UV region can be related to the unfolding of the enzyme molecule and exposure of aromatic amino acids [17]. The decrease of absorbance at Soret band requires modification of the heme or the apoprotein in a manner that attenuates the heme chromophore [18]. Jenzer et al [19] suggested OH'-induced oxidative cleveage of the porphyrin ring structure of the heine moiety of lactoperoxidase, resulting in the liberation of iron ions. They postulated such mechanism as a result of Fenton-like reaction, where, contrary to pulse radiolysis method, OH" can be formed near active site of the protein. As the shapes of visible absorption spectra of OH'-treated HRP solutions remain unchanged (data not shown) it seems that heme destruction does not occur under our reaction conditions. In order to check how OH'-induced protein modification influences compound I formation we investigated reaction (1) for HRP solutions 24 h after OH"

(%)

Loss of the ability of Loss of the abili~. compound I of substrate formation (%) binding (%) 0/0 0/0 0/15 0/21 26/45 41/48

0/0 0/0 23/13 34/19 27/19 28/23

action. We found that the rate constant of compound I formation is 2 x 107 M -! s-I and does not change after irradiation with a dose within the investigated range. The reaction yield, measured as [compound I]/[HRP], however, decreases with increasing [OH'] (with an initial lag in deoxygenated solutions) (table I). Inspection of the results collected in table I shows that activity loss is connected with such enzyme modification that causes both the interference with substrate binding and partial blocking of the channel used by peroxide. For samples treated with OH" in deoxygenated solutions, the peroxide channel is modified only when at least a ten-fold excess of OH" radicals over enzyme concentration is applied. As can be seen in table I, restrictions in the formation of compound I are higher in the presence of oxygen. The restrictions of substrate binding are slightly higher than in the case where HRP reacts with OH" in deoxygenated solution. Global loss of HRP activity is about 10% higher in the presence of oxygen.

Acknowledgments This work was supported in part by KBN grant no 6P20306006.

References 1 Davies KJA (1987) Protein damage and degradation by oxygen radicals. I. General aspects. J Biol Chem 262, 9895-9901 2 Gcbicka L, Metodiewa D (1988) Radiation-induced inactivation of bovine liver catalase in nitrous oxide-saturated so!utions. J Radioanal Nucl Chem Lett 127, 253-260 3 Gcbicka L, Gcbicki JL (1990) Pulse radiolysis of catalase in solution- II. Reactions of primary products from water radiolysis with catalase. ,rCadiat Phys Chem 36, 16 l - 163 4 Ggbicka L, Gcbicki JL (1992) Redox transformations in peroxidases studied by pulse radiolysis technique. Radiat Phys Chem 39, 113-116 5 0 h l s s o n PJ, Paul KG (1976) The molar absorptivity of horseradish peroxi. dase. Acta Chem Scand B30, 373-375 6 Karolczak S, Hodyr K, Lubis R, Kroh J (1986) Pulse radiolysis system based on ELU-6E LINAC. J Radioanal Nucl Chei,~ ! 0 !, 177-188

65 7 Karolczak S, Hodyr K, Po t~,uifiski M (1992) Pulse radiolysis system based on ELLI-6E LINAC. II. De~,iopment and upgrading the system. Radiat Pt~ys Chem 39, 1-5 8 Buxton GV, Greenstock f , , Helman WE Ross AB (1988) Critical review of rate constants for reactJ.~ns of hydrated electrons, hydrogen atoms and hydroxyl radicals (OH'/'y) in aqueous solution. J Phys Chem Ref Data 17, 513-886 9 Childs RE, Bardsley V,I,~2(1975) The steady-state kinetics of peroxidase with 2,2"-azino-di(3-ethyl-r~enzthiazoline-6-sulphonic acid) as chromogen. Bioehem J 145, 93-103 10 Asmus KD (1990) St Ifur-centered free radicals. In: Methods in Enzymology 186, 168-180 11 Armstrong DA (199IJ,.: Redox systems with sulfur-centered species. In: Sulfur-Centered Reacti,,'e Imermediates in Chemistry and Biology (Chatglialoglu C, Asmus KD, ed:~) Plenum Press, New York, 341-351 12 Schonbaum GR, Lo S (1972) Interaction of peroxidases with aromatic peracids and alkyl peroxides. J Biol Chern 247, 3353-3360 13 Bielski BHJ, Cabelli DE Arudi RL, Ross AB (1985) Reactivity of HO,JO,radicals in aqueous solutit.,n.J Phys Chem RefData 14, 1041-1100

14 Garrison WM (1987) Reaction mechanisms in lhe radiolysis of peptides, polipeptides and proteins. Chem Rev 87, 381-398 15 Das S, Mieden OJ, Pan X-M, Repas M, Schuchman MN, Schuchman H-E yon Sonntag C, Zegota H (1988) Aspects of the HO, elimination reaction from organic peroxyl radicals: some recent examples, in: Oxygen Radicals in Biology and Medicine (Simic MG, Taytor KA, Ward JF, yon Sonntag C, eds) Plenum Press, New York, 55-58 16 Metodiewa D, Ggbicka L, Bachman S (1987) The radiation-induced inactivation and post irradiation effects of HRP solution. J Radioanal Nucl Chem Lett i 19, 87-93 17 Dunford HB (1991) Horseradish peroxidase: Structure and kinetic properties. In: Peroxidases in Chemistry and Biology, vol !1 (Everse J, Everse KE, Grisham MB, eds) CRC Press, Boca Raton, 1-24 18 Ator MA, Oniz de Montellano PR (1987) Protein control of prosthetic heine reactivity. Reaction of substrates with the heine edge of HRP. J Biol Chem 262, 1542-1551 19 Jenzer H, Kohler H, Broger C (1987) The hydroxyl radicals in irreversible inactivation of lactoperoxidase by excess H,O~. Arch Biochem Biophys 258, 381-387