FreeRadical Biology&Medicine, Vol. 12, pp. 169-173, 1992 Printed in the USA. All fights reserved.

0891-5849]92 $5.00 + .00 Copyright © 1992 Pergamon Press ple

Brief Communication STABILITIES OF HYDROXYL RADICAL SPIN ADDUCTS OF PBN-TYPE SPIN TRAPS*

EDWARD G . JANZEN,~" YASHIGE KOTAKE and RANDALL D . H1NTON The National Biomedical Center for Spin Trapping and Free Radicals, Molecular Toxicology Research Program, Oklahoma Medical Research Foundation, 825 N.E. 13th Street, Oklahoma City, OK 73104, U.S.A.

(Received 26 July 1991; Accepted 8 October 1991) Abstract--The stability of the hydroxyl spin adduct of nine different PBN-type spin traps has been examined in phosphate buffer solutions of various pH. The hydroxyl adduct is produced by short illumination of hydrogen peroxide with UV light in the presence of spin trap and the decay of its EPR signal followed. The stability measured by the half life of the first-order decay is strongly dependent on the pH of the solution and the structure of the aromatic ring used in the trap. All hydroxyl adducts are more stable in acidic media, tert-Butyl hydroaminoxyl is detected as a degradation product of the hydroxyl adduct from all spin traps. Keywords--Hydroxyl radical, Spin trapping, PBN, EPR, Free radicals

INTRODUCTION

port on the first-order decay rates of the radical adducts of PBN-type spin traps in aqueous phosphate buffer solutions as a function of pH.

The hydroxyl radical is believed to be the most reactive species among oxygen radical intermediates which damage biological tissues. 1 When the hydroxyl radical is detected by the spin trapping technique2 the success of this detection method depends almost entirely on the stability of the hydroxyl radical spin adduct because the rate of formation of the hydroxyl adduct is very fast and does not depend significantly on the selected spin trap? -5 It is well recognized that the lack of basic data on the stability of oxyl adducts in spin trapping experiments limits the choice of spin traps when applied to biological systems. We have begun a screening program designed to provide some understanding of the decay processes of spin adducts in various environments and thereby to develop better spin traps for the detection of specific free radicals in biological systems. In this communication we re-

MATERIALS AND METHODS

The compounds used as spin traps were available from previous studies as indicated in references given in Table 1. Chemicals used for solution preparation (potassium dihydrogen phosphate buffer, hydrogen peroxide, deuterated water) are commercially available. A Fisher aqueous Minipet and a 10-in. 20 gauge stainless steel needle was used for the introduction of samples into the EPR spectrometer. The method used involves injecting premixed solutions of the spin trap (0.01 M), potassium phosphate buffer (0.05 M) and hydrogen peroxide (0.3 M) into the bottom of a flat cell positioned in the cavity of a BRUKER 300D EPR spectrometer. Since preliminary studies with PBN revealed that the concentrations of spin trap and H202 do not affect the firstorder rate constant all observations were performed using the above concentrations. A short period ofphotolysis (½s) from a Oriel 75W mercury lamp produced the hydroxyl adduct of the spin trap. Short photolysis periods were needed to avoid photochemical decomposition of the adduct. The decay of the third peak on the low field side of the six-line spectrum was followed

Address correspondence to Edward G. Janzen, Ph.D., The National Biomedical Center for Spin Trapping and Free Radicals, Molecular Toxicology Research Program, Oklahoma Medical Research Foundation, 825, N.E. 13th Street, Oklahoma City, OK 73104. * Publication No.2 from the National Biomedical Center for Spin Trapping and Free Radicals. t Alternate address: Departments of Clinical Studies and Biomedical Sciences, Ontario Veterinary College, University of Guelph, Guelph Ontario, Canada N IG2W 1. 169

170

E.G. JANZENel aL Table 1. HFS Constants of the Hydroxyl Adduct (Gauss) Spin Traps t

aN

aoHH

aoHH

/ - - - " k

( / ( ~ xx...~H-NI_..C(CH3)3 "~.~/ ~ ",'O

15.46 2.70

0.21

(PBN) 4-Pyridyl-N-t-butylnitrone (4-PyBN)

N~H=N_C(CH3)3~ O

15.17

1.88

0.30

3-Pyridyl-N-t-butylnitrone

~-~H=N_C(CH3)3

1~20

2.05

Phenyl-N-t-butylnitrone

(3-~BN)

2-Pyridyl-N-t-butylnitrone (2-PyBN)

N_~

~-~CH=N_C(CH3) 3 X__N (~

15.03 2.46

4-Pyridine-N-oxyl-N'-t-butylnitrone (4-PyOBN)

O--N~-CH=N-C(CH3)3x~/v O

15.03 1.67

3-Pyridine-N-oxyl-N'-t-butylnitrone (3-PyOBN)

~/-CH=N-C(CH3)3 /N*----' (~ O-

2-Pyridine-N-oxyl-N'-t-butylnitrone (2-PyOBN)

~H=N-C(CH3)3 \O-

14.94

0.32

1.60 0.26

15.47 2.73

4-N-Methylpyridyl-N'-t-butylnitrone (4-MPyBN)

H3C_N~_CH=N_C(CH3) ~ 3 14.82 O

1.41

0.41

3-N-Methylpyridyl-N'-t-butylnitrone (3-MPyBN)

~CH=N-C(CH3)3 /N *--J () H3C

1.43

0.36

(2-MPyBN)2"N'Methylpyridyl'N'-t-butylnitr°ne

~, _H~=/N - - C ( C ~)H 3 )

3

14.90

14.65 3.90

CH3 tNitrone spin traps are either commercially available or previously synthesized according to procedures indicated in the given references. 8,9

at 20°C by fixing the field position on this peak. Five half-lives were taken and averaged. Repeat experiments could be arranged by injecting new samples into the cell. The pH was adjusted by adding HC1 or KOH to pre-prepared solutions while monitoring the acidity with a pH meter. The assignment of the hydroxyl ~-hydrogen hyperfine splitting (hfs) was verified by using 96% deuterated water as solvent in some cases.

duces one set of three doublets presumably due to the hydroxyl radical adduct, providing the nitrone has the following partial structure and the ~-H hfs can be resolved:

+

- C H ~ - N - C ( C H 3 ) 3 + " O H --~ - C H - N - C ( C H 3 ) 3 . I

RESULTS AND DISCUSSION

Assignments The photolysis of low concentrations of hydrogen peroxide in the presence of nitrone spin traps pro-

OH i

O-

I

"O

At higher concentrations of hydrogen peroxide two sets of spectra result from trapping both the hydroxyl and the hydroperoxyl radicals6:

Stability of OH radical adducts

171

aHOH

OH H3C -

-Ell-

-C(CH3) 3 0

I

l0 gauss

Fig. 1. EPR spectrum of the hydroxyl adduct of 4-N-methylpyridyl-N'-tert-butyl nitrone produced by UV-iUumination of a phosphate buffer solution of hydrogen peroxide and the nitrone. Note the small hfs from the hydroxyl hydrogen on each line.

Kinetics

OOH +

I

- C H ~ N - C ( C H 3 ) 3 + "OOH --) - C H-N-C(CH3h. I

I

O-

"O

The assignments of some hydroxyl radical adduct spectra are based on the presence of an additional small y-H hyperfine splitting constant (hfs) which disappears in D 2 0 solutions. A typical EPR spectrum from the hydroxyl adduct of 4-N-methylpyridyl-N'tert-butyl nitrone is shown in Fig. 1. The additional splitting is assigned to the hydroxyl hydrogen. When the hfs from the hydroxyl hydrogen is not detected, assignment of the hydroxyl adduct spectrum has been based on the increase in intensity with increase in H202 (up to ca. 5%) in photogenerated systems and the relatively larger hfs from both the aminoxyl nitrogen and the [3-H, i.e., larger than the hfs from the hydroperoxyl adduct. However it should be noted that these assignments cannot be considered rigorous. The observed hfs's are listed in Table 1.

When the hydroxyl adduct is generated at relatively high concentrations the initial decay of the spectrum assigned to the hydroxyl adduct is second order due to disproportionation between two spin adducts. When the concentration is decreased to less than approximately 1 x 10-5 mol/l the decay kinetics change to first order. The nature of the complete decay profile of the spin adducts observed varies depending on the spin trap selected. At low pH often a second spectrum with very similar hyperfine splitting constants is observed which decays much more slowly and sometimes survives for a long time. The structure of this second species is not known at this time. Only those spectra which provide the additional hfs due to the hydroxyl hydrogen are considered to be dependably assigned. When the stable secondary adduct was produced the decay curve does not completely return to the baseline thus causing an error in calculating the decay rate constants. For the whole pH range investigated the hydroxyl adduct of all spin traps studied

UV flash

×

O

O

Fig. 2. Time course of the rise and decay of the aminoxyl radical after UV illumination of phosphate buffer solution (pH 7.4) of hydrogen peroxide and 2-N-methypyridyl N-tert-butyl nitrone. The lines marked by O are from the hydroxyl adduct; the lines marked by X are from tert-butyl hydroaminoxyl. Total scan time is 84 s over 50 Gauss.

172

E.G. JANZENel al.

produces tert-butyl hydroaminoxyl after the end of illumination (Fig. 2). In this study we give first-order decay rates constants of only the hydroxyl adducts where the decay profile shows a linear decrease in a logarithmic plot of signal intensity; however, the assignments must sometimes be made without the presence of the hydroxyl hydrogen splitting since not all spin adducts give spectra with this feature.

Specific spin traps The additional doublet splitting observed in some hydroxyl adducts disappears in D20. Thus the splitting is assigned to the hydroxyl 7-hydrogen hfs. This splitting can still be detected in basic solutions. The initial EPR spectra of 3- and 4-MPyBN but not 2MPyBN give similar small splittings assigned to the hydroxyl ~'-hydrogen hfs of the spin adducts of the hydroxyl radical at pH 6.0 (see Table 1). The decay rates of these initial spectra as indicated by decay half lives vary considerably (Table 2). Small hyperfine splittings are also detected for the 3- and 4-PyOBN spin adducts of the hydroxyl radical although not for 2-PyOBN. Surprisingly the lifetime of the spin adduct of the 2-isomer is very long, some 8 times that of PBN at pH 6.0 (Table 2). For the hydroxyl spin adduct of the 4-N-methylpyridinium tert-butyl nitrone similar results are obtained, namely the 4-MPyBN adduct is as persistent as the adduct from 4-PyBN and 4PyOBN. However, the adduct from 2-MPyBN is more stable than that from 2-PyBN but less persistent than the adduct from 2-PyOBN (Table 1). The hydroxyl spin adduct stabilities (at pH 7.4) follow the sequence 2-PyOBN ~ 4-PyBN > 3-PyBN > 4MPyBN > others.

in pH. The decay rate constants reach a plateau in solutions of extreme pH. In fact the plots of observed lifetimes of the spin adducts as a function ofpH resemble a titration curve for an acidic proton (Fig. 3) which is well reproduced by assuming rate constants in acidic and basic media and an ionization constant. 7 The implication is that the hydroxyl hydrogen is removed in the decay process. A mechanism which would be consistent with the observed first-order decay and the pH effect is cleavage of the nitrogen-carbon bond concomitant with hydroxyl hydrogen removal to produce benzaldehyde and the anion radical of 2-methyl-2-nitrosopropane: OH - C H - -C(CH3) 3 + - O H O" O-

~

- / H - N - C ( C H 3 ) 3 + H20

O" O-

O

-ca-

-C(CH3) 3 ~

-

H +-N-C(CH3)3.

I

O"

O"

The anion radical of 2-methyl-2-nitrosoprotonate would protonate rapidly to give the tert-butyl hydroaminoxyl in every case: -N-C(CH3) 3 + H20 ~ L

H-N-C(CH3) 3 + -OH I

O"

O"

Effect of pH and mechanism of decay

SUMMARY

The effect of pH on the stability of the hydroxyl spin adducts is to decrease the lifetime with increase

In summary, the stability of the hydroxyl spin adducts of PBN-type spin traps varies by about 40 times

Table 2. Half Life (First-Order) of Hydroxyl Spin Adducts as a Function of pH in Aqueous Phosphate Buffer Spin Traps

Phenyl-N-t-butylnitrone 4-PyridyI-N-t-butylnitrone 3-Pyridyl-N-t-butylnitrone 2-Pyridyl-N-t-butylnitrone 4-Pyridine-N-oxyl-N'-t-butylnitrone 3-Pyridine-N-oxyl-N'-t-butylnitrone 2-Pyridine-N-oxyl-N'-t-butylnitrone 4-N-Methyipyridyl-N'-t-butylnitrone 3-N-Methylpyridyl-N'-t-butylnitrone 2-N-Methylpyridyl-N'-t-butylnitrone

Half Life/s

PBN 4-PyBN 3-PyBN 2-PyBN 4-PyOBN 3-PyOBN 2-PyOBN 4-MPyBN 3-MPyBN 2-MPyBN

pH 6.0

pH 7.4

pH 8.0

90 185 120 23 80 56 730 430 240 290

38 31 28 24 10 22 610 120 24 30

11

3 5 3 3 2 43 13 5 2

Stability of OH radical adducts

173

Resources in the National Institute of Health. This research was supported by grant No. RRO 5517-01A1. t~/s400300 4 - M P y B N ~

REFERENCES

I

6

7

8

pH

9

Fig. 3. First order decay half-lives of the hydroxyl adduct of some spin traps plotted as a function of pH of the solution.

at pH 6.0 or 7.4 and about 20 times at pH 8.0. The hydroxyl spin adduct of 2-PyOBN is the most persistent at all pHs studied. No obvious trend can be recognized within the series of nitrones studied. Further investigations on 2-PyOBN spin adducts are underway. Acknowledgement - - The National Biomedical Center for Spin

Trapping and Free Radicals is supported by the Biomedical Research Technology Program of the National Center for Research

1. Packer, L.; Glazer, A.N., eds. Oxygen radicals in biological systems. Part B Oxygen Radicals and Antioxidants. Methods in enzymology. Vol. 186. New York: Academic Press; 1990. 2. Janzen, E.G.; Haire, D.L. Two decades of spin trapping. In: Advances in free radical chemistry. Vol. 1. Greenwich, CN: JAI Press; 1990:253-295. 3. Finkelstein, E.; Rosen, G.M.; Raukman, E.J. Spin trapping. Kinetics of the reaction of superoxide and hydroxyl radicals with nitrones. J. Am. Chem. Soc. 102:4994-4999; 1980. 4. Marriot, P.R.; Perkins, M.J.; Griller, D. Spin trapping for hydroxyl in water: A kinetic evaluation of two popular traps. Can. J. Chem. 58:803-807; 1980. 5. Sridhar, R.; Beaumont, P.C.; Powers, E.L. Fast kinetics of the reactions of hydroxyl radicals with nitrone spin traps. J. Radioanal. Nucl. Chem. 101:227-237; 1986. 6. Harbour, J.R.; Chow, V.; Bolton, J.R. An electron spin resonance study of the spin adducts of OH and HO2 radicals with nitrones in the ultraviolet photolysis of aqueous hydrogen peroxide solutions. Can. J. Chem. 52:3549-3553; 1974. 7. For example, see Fersht, A. Enzyme structure and mechanism. San Francisco, CA: W.H. Freeman and Co.; 1977. 8. Janzen, E.G.; Dudley, R.L.; Shetty, R.V. Synthesis and electron spin resonance chemistry of nitronyl labels for spin trapping. a-phenyl-N-[5-(5-methyl-2,2-diaikyl-1,3Mioxanyl)]nitrones and ct-(N-alkylpyridinium)N-tert-butyl nitrones. J. Am. Chem. Soc. 101:243; 1979. 9. Janzen, E.G.; Wang, Y.Y.; Shetty, R.V. Spin trapping with apyridyl-N-oxide N-tert-butyl nitrones in aqueous solutions. A unique ESR spectrum for the hydroxyl radical adduct. J. Am. Chem. Soc. 100:2923; 1978.