Thiol-induced hydroxyl radical formation and scavenger effect of thiocarbamides on hydroxyl radicals

Thiol-Induced Hydroxyl Radical Formation and Scavenger Effect of Thiocarbamides on Hydroxyl Radicals Noriko Motohashi and Itsuhiko Mori Department of Pharmaceutical Radiochemistry, Kobe Women’s College of Pharmacy, Kobe, Japan

ABSTRACT The effects of thiols and thiocarbamides on hydroxyl radical (*OH) formation by the hypoxanthine(HYP)-xanthine oxidase(XOD)-Fe ‘+ *EDTA system were investigated in the range of 0.5-S mM by calorimetrically measuring salicylate hydroxylation. Thiocarbamides powerfully inhibited the hydroxylation while thiols showed a paradoxical effect, enhancing it at low concentrations, but inhibiting it at high ones. Thiols in the presence of Fef+ .EDTA generated superoxide anions (Of) and *OH during the oxidation, but thiocarbamides did not. A study of the effect of ergothioneine, a thiccarbamide present in mammals, on the *OH spin adduct of 5,5dimethyl-1-pyrroline-N-oxide(DMPO) by EPR spectrometry showed that it effectively decreased the *OH spin adduct without causing the appearance of other signals. Reaction mechanisms are proposed for the 0; evolution and *OH formation by the thiols themselves in the presence of Fe)+ *EDTA and *OH with thiols and thiocarbamides.

INTRODUCTION The highly reactive hydroxyl radicals (-OH), which may be the true damaging species in vivo, are produced via superoxide anions (OF) by the iron-catalyzed Haber-Weiss reaction. The reaction proceeds more rapidly in the presence of EDTA [l]. The autoxidation .of dithiothreitol, 2-mercaptoethanol, ethylmercaptan, glutathione(GSH) and cysteine produces OF, -OH, and thiyl radicals [2-4]. Furthermore, cysteine and GSH in the OF-generating enzyme system enhance or inhibit *OH formation depending on their concentrations [5]. Thiourea(thiocarbamide) has been shown to be more effective a *OH scavenger than mannitol and formic acid [l]. It also completely abolishes *OH spin adducts [6].

Address reprint requests to Dr. Nor&o Motohashi, Department of Pharmaceutical Kadiochemistry, College of Pharmacy, Motoyamakita-machi, Kobe 658, Japan.

Journal of Inorganic Biochemistry 26,205-212 (1986) 0 1986 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Ave., New York, NY 10017

Kobe Women’s

205 0162-0134/86/$3.50

206

N. Motohashi and I. Mori

Recently, N-methyl-2-mercaptoimidazole (MMI, methimazole), which has a thiocarbamide moiety, has been found to directly interact with -OH and protect sensitive enzymes against inactivation [7]. We have shown that ergothioneine (ESH), the only than mercaptoimidazole derivative in nature, is a more effective radioprotector cysteine and cysteamine [S]. Reported here are the different effects of thiols and thiocarbamides on *OH formation and scavenging as well as their proposed reaction mechanisms. MATERIALS

AND METHODS

Superoxide dismutase(SOD), catalase, and ESH were obtained from Sigma. 2Mercaptoimidazole(M1) and MMI, purchased from K & K Laboratories, were used after recrystallization from ethanol. o-Mercaptopropionylglycine (a-MPG) was a gift from Santen Seiyaku (Japan). Coenzyme A(CoA) was obtained from Kojin (Japan). Xanthine oxidase(XOD) was purchased from Boehringer. Ferric EDTA was obtained from Dojindo (Japan). The spin trap 5,5-dimethyl-1-pyroline-N-oxide(DMP0) was purchased from Aldrich Chemicals. All other reagents were of the highest quality obtainable. Calorimetric assay of *OH formation was performed by measuring salicylate hydroxylation [9] in the presence and absence of the hypoxanthine(HYP)-XOD system. The 0; production was assayed at 550 nm by reducing nitroblue tetrazolium (NBT) to formazan at 25°C. EPR spectra were recorded with a field set 3370 G, modulation frequency 100 kHz, modulation amplitude 1.0 G, and microwave frequency 9.442 GHz using a JES-FE-3X spectorometer. RESULTS The calorimetry of hydroxylation was performed in the presence of 100 PM Fe 3+ - EDTA because it induces about twice the hydroxylation compared to FeS04 and FeC13 [lo]. Table 1 shows the effects of thiols and thiocarbamides on the salicylate 3+ *EDTA system. Thiols at 0.5 mM, except for hydroxylation by the HYP-XOD-Fe GSH and CoA, enhanced the *OH formation. Cysteamine and cysteine at 1 mM further increased the hydroxylation. Even cysteamine at 5 mM did not inhibit it. However, CoA, which consists of cysteamine, pantothenic acid, and phosphoadenosine pyrophosphate, decreased the hydroxylation at the concentration used. As CoA at 5 mM slightly inhibited the enzyme activity, the inhibition by 5 mM CoA should be less than the present data. Thiocarbamides, such as ESH, MMI, MI, and thiourea, effectively decreased the *OH formation at the concentration used and at 5 mM displayed 80%-90% inhibition. None of the sulfur compounds used except CoA significantly inhibited the enzyme activity. The inhibition by mannitol was about onethird of that by thiourea. The possibility of 0; scavenging by thiocarbamides was investigated using the NBT-HYP-XOD system because of their powerful inhibitory effect on the hydroxylation. Table 2 shows the initial rate of formazan formation in the presence of thiocarbamides together with the data of cysteamine and cysteine. Thiocarbamides did not inhibit the NBT reduction at all, while cysteamine and cysteine slightly increased the initial rate and with time enhanced the formazan formation (not shown). These results suggest that the powerful inhibition by thiocarbamides of salicylate hydroxylation is not due to 0; scavenging.

207

Thiols, Thiocarbamides, and Hydroxyl Radicals

f~

0

g-r:

I

e-

< [-, m

I

0

a&

0

011 0

0

•~

°~

~

0

0

I l l

I

.o •~ 0

~o~ 0

= ~'~

m

o

~

~

~

208

N. Motohashi and I. Mori

I I

Thiols, Tbiocarbamides, and Hydroxyl Radicals

B 0

209

. . . . 1.0 2.0 3.0 4.0 5.0 Thiol concentratton(mU)

FIGURE 1. Production of hydroxyl radical by various concentrations of cysteamine (O-O) and cysteine ( l * * *0) in the presence of Fe3+ *EDTA. Reaction mixtures were as described in Table 1 without hypoxanthine and xanthine oxidase.

The stimulatory effect of several thiols on the hydroxylation as shown in Table 1 might mean that thiols themselves generate -OH at significant rates through the reaction. The salicylate hydroxylation by sulfur compounds themselves in the presence of Fe3+ *EDTA was investigated without the HYP-XOD system. None of the thiocarbamides at the concentration used led to significant hydroxylation. The maximum amounts of hydroxylation by cysteamine and cysteine were 230 and 140 nmol, respectively (Fig. 1). Hydroxylation by other thiols amounted to lo-30 nmols at the concentration used. In the absence of Fe 3+ *EDTA, scarcely any hydroxylation was found. These facts indicate that thiols themselves significantly generate -OH through the reaction with Fe3+ *EDTA. The *OH generation by thiols themselves suggests that 0; should also be generated during the reaction. 0; evolution by cysteamine itself was investigated using NBT reduction (Fig. 2). Cysteamine effectively reduced NBT in the presence of Fe3+ *EDTA (Fig. 2c), but scarcely in the absence of Fe3+*EDTA (Fig. 2f). Accordingly, NBT should not be directly reduced by ionized cysteamine. SOD stimulated the formazan formation (Fig. 2a) whereas catalase inhibited it (Fig. 2d). These results show that hydrogen peroxide, which is produced as a result of 0; dismutation, accelerates the NBT reduction. The inhibitory effect of DMPO (Figs. 2b and 2e) suggests that *OH and the thiyl radical contribute to the over-all rate of NBT reduction. The -OH scavenger effect of ESH was measured by EPR spectrometty (Fig. 3). The spectrum (Fig. 3a) showed the characteristic 1:2:2:1 quartet pattern and the parameters ON = an = 14.7 G for the *OH spin adduct of DMPO(DMPO-OH) [ll, 121. ESH at 6 mM (Fig. 3b) reduced the amount of DMPO-OH. No other signals were observed.

210

N. Motohashiand I. Mori

a

0

2

4

6 8 Time(m3.n)

3.0

FIGURE 2. Effect of Fe3+ .EDTA, SOD, catalase, and DMPO on the NBT reduction by cysteamine. Reaction mixtures contained 500 PM NBT, 0.17 % gelatin, 5.0 mM cysteamine, and 100 pM Fe’ + . EDTA in a total volume of 3 ml 0.15 M phosphate buffer (pH 7.4). (a) SOD 290 units/ml, (b) SOD 290 units/ml and DMPO 100 mM, (c) control, (d) catalase 100 units/ml, (e) DMPO 100 mM, (f) no Fe’+*EDTA present. FIGURE 3. Effect of ergothioneine on hydroxyl radical trapping by DMPO. Reaction mixtures contained (a) control (DMPO-OH): 0.12 56, H202, 40 pM FeS04, 87.5 mM DMPO in 50 mM phosphate buffer, pH 7.4; (b) the control mixture plus 6 mM ergothioneine. EPR details were as described in Materials and Methods.

211

Thiols, Thiocarbamides, and Hydroxyl Radicals

DISCUSSION

The mechanism of 0; and *OH generation by the reaction of thiol with Fe3 + *EDTA can be expressed by the following series of reactions: RS+Fe3+*EDTA-+RS*

+Fe2+*EDTA

RS. +RS- +02-rRSSR+O;

(1) (2)

Fe3+*EDTA+O;+Fe2+*EDTA+02

(3)

Fe2+*EDTA+02+Fe3+*EDTA+Oi

(4)

0;+0;+2H++H202+02 RS. +RS- +H,02-‘RSSR+

(5) .OH+OH-

Hz02 + Fez+ .EDTA+Fe3+ .EDTA + *OH + OH RS- + *OH-rRS. +OHRS. +RS.-,RSSR

(6) (7) (8)

(9) Thiol(RS-) reduces Fe3+ .EDTA to Fe 2+ - EDTA and is oxidized to the thiyl radical (RS.) (reaction (1)). 0; is generated by the reduction of O2 by RS. and RS- (reaction (2)). Fe2+ *EDTA also arises from the reaction of Fe3 + *EDTA with 0; (reaction (3)). The dismutation of 0; produces H202 (reaction (5)), which reacts with RS* and RS- to form *OH (reaction (6)). 0; and *OH are also generated by the oxidation of Fe2+ *EDTA (reaction (4) and (7)). The powerful oxidant *OH is scavenged by the formation of disulfide (reactions (8) and (9)). The OF-generating enzyme system accelerates reactions (3) and (5). The paradoxical effect of thiols on - OH would be due to the competition between reactions (l)-(7) and reactions (8) and (9). The NBT reduction by cysteamine shown in Figure 2 is anticipated as resulting from the contribution of OF, RS., and *OH. The stimulation by SOD and the inhibition by catalase are attributed to acceleration of reaction (5) and depression of reactions (6) and (7), respectively. Accumulation of -OH accelerates reaction (8). DMPO serves to diminish ‘OH and RS. in the overall reaction. However, 0; is not significantly reduced by DMPO because of its low reactivity with DMPO [12]. Therefore, NBT .reduction in the presence of thiol and Fe 3+ *EDTA would be caused by 0; and directly by RS *. No stimulatory effect of thiocarbamides on the -OH formation can be explained by the difference of redox potentials (E,‘) between thiols and thiocarbamides. Ei for thiols lies generally in the range of - 0.20 to - 0.40 V, while that for ESH is - 0.06 V [ 131. Therefore, thiocarbamides can scarcely promote the reduction of Fe3 + - EDTA (reaction (l)), which initiates the reactions of *OH formation. Furthermore, the powerful inhibition by thiocarbamides might be due to their thione form. ESH in aqueous solution is predominantly present in the thione form [14], which is resistant against the autoxidation. The disulfide is unstable in water and is reduced immediately to the Gone form [ 151. Consequently, the regenerated thiocarbamide reacts with -OH again. Lowering the *OH spin adduct concentration of DMPO by ESH would depend on the rate constant for the reaction of ESH and -OH. The rate constant for DMPO + OH (k = 2.1 x 109M-’ set-‘) [12] is lower than those for MM1 + *OH (k = 1.36 x lOi M-i set-i) [7] and for thiourea + -OH (k = 4.7 x lo9 M-’ set-‘) 1161.

212

N. Motohashi

and I. Mori

MM1 is presumed to react directly with *OH [7]. Accordingly, thiocarbamides seem to rapidly react with *OH and scavenge -OH. The EPR spectrum of DMPO in the presence of ESH (Fig. 3b) shows only the reduced *OH spin adduct of DMPO. Thus, the generation of thiyl radicals of ESH is unlikely to occur during the reaction of ESH and *OH. ESH is distributed widely in mammalian tissues at about 40-400 nmol per gram of fresh tissue [17]. It might act as a radical scavenger in pathological systems.

REFERENCES 1. B. Halliwell, FEBS Left. 92, 321-326 (1978). 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

H. P. Misra, J. Biol. Chem. 249, 2151-2155 (1974). A. I. F. Searle and A. Tomasi, J. fnorg. Biochem. 17, 161-166 (1982). Cl. Saez, P. J. Thornally, H. A. 0. Hill, R. Hems, and J. V. Bannister, Biochim. Biophys. Acta 719, 24-31 (1982). D. A. Rowley, FEBS Letr. 138, 33-36 (1982). R. A. Floyd and C. A. Lewis, Biochemistry 22, 2645-2649 (1983). J. J. Taylor, R. L. Willson, and P. Kendall-Taylor, FEBS Lett. 176, 337-340 (1984). N. Motohashi, I. Mori, Y. Sugiura, and H. Tanaka, Chem. Pharm. Bull. 25, 2516-2523 (1977). R. Richmond, B. Halliwell, J. Chauhan, and A. Darbre, Anal. Biochem. 118, 328-335 (1981). N. Motohashi and I. Mori, FEBS Left. 157, 197-199 (1983). J. R. Harbour, V. Chow, and J. R. Bolton, Can. J. Chem. 52, 3549-3553 (1974). E. Finkelstein, G. M. Rosen, and E. J. Rauckman, J. Am. Chem. Sot.. 102, 4994-4999 (1980). P. C. Jocelyn, Biochemistry of the SH group, Academic, New York, 1972, pp. 55-56. N. Motohashi, I. Mori, and Y. Sugiura, Chem. Phurm. Bull. 24, 1737-1741 (1976). H. Heath and G. Toennies, Biochem. J. 68, 204-210 (1958). M. Anbar and P. Neta, Int. J. Appl. Radiut. 18, 495-523 (1967). P. C. Jocelyn, Biochemistry of the SH group, Academic, New York, 1972, p. 10. Received

July 25, 1985; accepted November

4, I985