Selective inhibition of Fe- versus Cu/Zn-superoxide dismutases by 2,3-dihydroxybenzoic acid derivatives

Article abstract of DOI:10.1248/cpb.50.578

A series of catechol derivatives were synthesised and tested for their ability to inactivate the iron-containing superoxide dismutase (Fe-SOD) from Escherichia coli and the bovine erythrocytes Cu/Zn-SOD. Incubation of catechols with Fe- or Cu/Zn SODs resu

Full text of DOI:10.1248/cpb.50.578

578
Chem. Pharm. Bull. 50(5) 578—582 (2002)
Vol. 50, No. 5
Selective Inhibition of Fe- versus Cu/Zn-Superoxide Dismutases by
2,3-Dihydroxybenzoic Acid Derivatives
Laurent SOULÈRE, Cécile VIODÉ, Jacques PÉRIÉ, and Pascal HOFFMANN*
Groupe de Chimie Organique Biologique, Laboratoire de Synthèse et Physico-Chimie de Molécules d’Intérêt Biologique;
UMR-CNRS 5068-Université Paul Sabatier, 31062 Toulouse cedex, France.
Received September 21, 2001; accepted January 15, 2002
A series of catechol derivatives were synthesised and tested for their ability to inactivate the iron-containing
superoxide dismutase (Fe-SOD) from Escherichia coli and the bovine erythrocytes Cu/Zn-SOD. Incubation of
catechols with Fe- or Cu/Zn SODs resulted in a time-dependent loss of enzyme activity with highly selective inhi-
bition for the iron-dependent enzyme. Catechol-induced inactivation of SODs was correlated with the auto-oxi-
dation of the catechol compounds to their corresponding ortho-quinone derivatives, which was found to be non-
dependent on the presence of enzymes. Mass electrospray experiments on catechol-incubated Fe-SOD provided
evidence for the irreversible nature of the inhibition process, yielding to a complex mixture of modified proteins.
Key words catechol; quinone; Fe-SOD; Cu/Zn-SOD; inhibitor
Metalloenzymes superoxide dismutases (SODs) are key Results
enzymes in the metabolism of oxygen free radicals whose
Synthesis of Catechol Derivatives (Chart 1) Commer-
primary activity is the dismutation of superoxide radical cially available 2,3-dihydroxybenzoic acid 1 was used as
anion into hydrogen peroxide and molecular oxygen.1) The starting material for the synthesis of all catechol derivatives
iron-dependent superoxide dismutases (Fe-SODs), struc- 2—8. Catechol 2 was prepared from 1 and [(tert-butyloxy-
turally distinct from the Cu/Zn family of SODs and, to a carbonyl)amino]ethylamine8) with 1,3-dicyclohexylcarbodi-
lesser extent, from the manganese family of SODs (Mn- imide (DCC) as coupling reagent. Removal of the butyloxy
SODs),2) constitute in a number of parasitic protozoan orga- carbonyl protecting group (Boc) from 2 under acidic condi-
nismsthe primary enzymatic defence against damage caused tions using trifluoroacetic acid gave hydrochloride compound
Ϫ
·
by the superoxide anion radical O2, and its derivatives such 3. Compounds 4 and 6 were obtained from 2,3-dihydroxy-
as peroxynitrite. As a consequence, it has been proposed that benzoic acid using the DCC method in the presence of
the development of specific inhibitors of parasitic iron super- glycine ethyl ester or N-hydroxysuccinimide, respectively.
oxide dismutases, which are different from their host’s Cu/Zn Treatment of 4 in aqueous sodium hydroxide and subsequent
or Mn-dependent enzymes, may lead to the perturbation of acidification gave compound 5, and reaction of 6 with 1,2-di-
the antioxidant defence mechanism, and thus could con- aminoethane and 1,6-diaminohexane yielded the bis-catechol
tribute to the development of new drugs with anti-parasitic derivatives 7 and 8, respectively. All compounds were then
properties.3) The bis-catechol derivative N1,N6-bis(dihydroxy- fully characterised by mass spectrometry, 1H-, 13C-NMR and
benzoyl)-1,6-diaminohexane 8 (Chart 1) has been shown to IR spectroscopies.
be a selective and irreversible inhibitor of the iron-SOD from
Oxidation Kinetic Measurements The oxidation of cate-
Crithidia fascicula, a non-parasitic trypanosomatid related to chols to their corresponding ortho-quinone derivatives was
trypanosomes.4) The authors showed that inhibition was determined by UV-visible spectroscopy by monitoring the in-
caused by a mechanism other than chelation, and hypothe- crease in absorbance at lϭ280 nm. As shown in Fig. 1A, au-
sized that this bis-catechol was oxidized to a quinone before toxidation of compound 5 at 400 mM occurred through a very
forming a covalent bond with the protein. In addition to any slow process in a 50 mM Tris buffer (pH 8.2) at 25 °C, with a
inhibitory effects, there is an interest in studying the reacti- first order constant value of 0.079 hϪ1 (t1/2ϭ9 h). The addi-
vity of catechol derivatives towards biological molecules tion of either Escherichia coli Fe-SOD or bovine erythro-
such as proteins, whose structural modifications are often asso- cytes Cu/Zn-SOD had no significant effect on the oxidation
ciated in the development of many diseases and in aging. In- rate with half-life time values of about 11 and 10 h, respec-
deed, a number of drugs can undergo oxido-reductive trans- tively (Fig. 1A). A similar effect was observed on all other
formations in cells, such as O-demethylation reactions cata- compounds (data not shown), thus providing proof that this
lyzed by cytochrome P450-dependent mono-oxygenases, or class of enzyme neither catalyses nor inhibits the conversion
one- or two-electron oxidation mediated by peroxidases and of catechol compounds to their corresponding quinones. In
tyrosinases, leading to the formation of catechol and/or these experiments, it should be noted that catechols (400 mM)
ortho-quinone moieties as major metabolites.5—7)
were in large excess compared to enzyme amount (0.3 mM),
In the present work, as part of a program to develop novel explaining why an inhibition of auto-oxidation rate, which
iron-superoxide dismutase inhibitors, we synthesised and could be due to substrate consumption by enzyme through
evaluated a series of catechol derivatives. The influence of covalent bonding, was not observed. In comparison, when ty-
added SOD on the rate of auto-oxidation reaction of these rosinase, known to catalyse the production of ortho-quinones
compounds to their corresponding ortho-quinones was also from phenol and catechol compounds, was added, oxidation
investigated.
of compound 5 occurred much more rapidly (Fig. 1B).
Inhibition Studies The activities of catechol-incubated
To whom correspondence should be addressed. e-mail: hoffmann@cict.fr
© 2002 Pharmaceutical Society of Japan

May 2002
579
(a) 2: [(tert-Butyloxycarbonyl)amino]ethylamine, DCC, THF, 36%; 4: glycine ethyl ester hydrochloride, Et3N, DCC, THF, 40%; 6: N-hydroxysuccinimide, DCC, THF, 36%. (b)
3: TFA, CH2Cl2, 72%; 5: 1 M aq. NaOH, 84%; 7: ethylene diamine, THF, 0 °C, 41%; 8: 1,6-hexane diamine, THF, 0 °C, 35%.
Chart 1
Fig. 1. (A) Effect of E. coli Fe-SOD (0.3 mM) and Bovine Erythrocyte Cu/Zn-SOD (0.3 mM) on the Auto-oxidation of Compound 5 (400 mM) and (B) Effect
of Tyrosinase (5 U/ml) on the Auto-oxidation Initial Rate of Compound 5 (400 mM)
All reactions were performed in a 50 mM Tris buffer pH 8.2, and spectrophometrically monitored at 280 nm.
Fig. 2. Dose-Dependence of the Inhibition of E. coli Fe-SOD and Bovine Erythrocyte Cu/Zn-SOD by Catechol Derivatives 2—5, 7 and 8 after 24 h at
25 °C in a 50 mM Tris Buffer pH 8.2 with 1 mM EDTA
Superoxide dismutase activities were determined using a standard SOD assay based on the auto-oxidation of pyrogallol.
SODs were determined using the SOD assay based on pyro- catechol-induced inactivation is an irreversible process. For
gallol auto-oxidation in a 50 mM Tris buffer (pH 8.2) contain- compounds 3 and 5, the t1/2 values (time corresponding to
ing 1 mM EDTA. As shown in Fig. 2, all compounds of this 50% inhibition) were found to be 25 min and 11 h, respec-
study demonstrated a selective inhibitory effect on the iron- tively. However, the corresponding kinetic data (Ki and ki or
dependent enzyme in a dose-dependent manner. Except for kinact) could not be determined due to the uncertainty of
compound 2, all catechol derivatives at concentrations of ortho-quinone concentrations, which are generated in situ by
400—500 mM fully inhibited the activity of the iron-depen- spontaneous catechol oxidation, and which are known to de-
dent enzyme, whereas bovine erythrocyte Cu/Zn-SOD was compose rapidly in aqueous solution. In the attempt to iden-
slightly inactivated at these concentrations (with the only ex- tify the modifications induced by catechol derivatives, E. coli
ception being for compound 3). Inhibition by all compounds Fe-SOD (30 mM) was incubated with compound 5 (40 mM)
(400 mM) was found to be time-dependent, suggesting that for 24 h, extensively dialysed against water, and analysed by

580
Vol. 50, No. 5
Chart 2. Comparative Schematic Views of the Shape of Active Site Channels of E. coli Fe-SOD (A) and Bovine Erythrocyte Cu/Zn-SOD (B)
The data files were obtained from the Protein Data Bank at Brookhaven National Laboratory (PDB access codes for the coordinate sets are E. coli Fe-SOD, 1ISA and bovine ery-
throcyte Cu/Zn-SOD, 1CBJ).
electrospray mass spectrometry. After this treatment, native SOD. However, this method did not provide information re-
SOD (apo subunit 21136 Da) was not recovered on the mass garding the nature of the residues associated with the irre-
spectrum which displayed a complex molecular mass distri- versible process. It can be assumed that cysteines are more
bution, corresponding to calculated deconvoluted masses up susceptible to ortho-quinone-induced modifications,16) but
to around 23500 Da, suggesting extensive product hetero- the large number of adducts formed relative to the single
geneity.
available cysteine residue per subunit requires that other nu-
cleophilic amino acid residues, such as lysines or histidines,
also be modified. In all known Fe-SODs, the iron is coordi-
Discussion
The primary activity of superoxide dismutases is the dis- nated to three histidines, one aspartate and a solvent mole-
mutation of superoxide radical anion into hydrogen peroxide cule as fifth ligand, forming a trigonal bipyramid around the
and molecular oxygen. However, other interesting catalytic metal ion.17) These four inaccessible ligand residues are
properties comprise the ability of SODs to either increase or highly conserved, as are His30 and Tyr34 located at the ver-
decrease the rate of some oxidation reactions. For instance, tex of the funnel, which block access to the active site of the
the auto-oxidation of pyrogallol,9) epinephrine10) and 6-hy- enzyme. Consequently, it seems unlikely that one residue of
droxydopamine11) is inhibited by SODs because superoxide the active site could be responsible for the nucleophilic at-
anion acts as a chain propagation species. In contrast, when tacks leading to irreversible modifications of the protein.
superoxide is the product of a reversible process, autoxida- Several studies, based on chemical modification and site-di-
tion rates can be enhanced in the presence of SODs. This be- rected mutagenesis on charged residues of SODs, showed the
haviour, for example, is observed in the auto-oxidation of 3- existence of an electrostatic control of substrate diffusion in-
hydroxyanthranilic acid to cinnabarinate.12) Conversely, the volving positively charged residues.18—22) It can be assumed
auto-oxidation of 1,4-naphthohydroquinone has been shown that the catechol-induced loss of activity of E. coli Fe-SOD is
to occur rapidly in a phosphate buffer (pH 7.4), and the addi- attributable to covalent modifications occurring at these
tion of Mn- or Cu/Zn-SOD has no effect on the course of its residues located at the entrance of the channel conducting to-
auto-oxidation, except a slight increase in the lag phase.13) In wards the active metal ion (i.e. Lys 29, Lys 116 or Arg 170;
recent studies, it has been shown that the rate of auto-oxida- Chart 2A). In particular, Arg 170, which is conserved in the
tion of some substituted naphthohydroquinone derivatives, large majority of Fe-SOD sequences, and which has been
including 2,3-dimethyl-1,4-naphthohydroquinone, decreases shown to be critical for catalytic activity,23,24) could be one of
with increasing concentration of bovine erythrocytes Cu/Zn the major target residues. The difference that we observed in
SOD,14,15) but the same enzyme has no effect on the rate the inhibition of the two classes of enzymes by the oxidation
of auto-oxidation of some halogenated 1,4-naphthohydro- products of catechols should be related to binding at this spe-
quinones, while that of the corresponding 5-hydroxy deriva- cific arginine residue, which is likely to be more accessible in
tive is increased.15) In our context, we observed that E. coli the Fe-SOD (Arg 170) than in the Cu/Zn-SOD (Arg 141)
Fe-SOD and Cu/Zn-SOD had little or no effect on the rate of (Chart 2).
the auto-oxidation of catechol derivatives to their corres-
ponding quinones. The kinetic data indicated a time-de-
Experimental
General Nuclear magnetic resonance spectra were recorded on a Bruker
pendent inactivation, suggesting an irreversible inhibition
AC 250 at 250 MHz (1H) or 60 MHz (13C), ultraviolet spectra with a UV mc2
process of superoxide dismutases. To confirm the covalent
spectrophotometer (SAFAS), and IR spectra on a Perkin-Elmer 1600 FTIR
modification on Fe-SOD by catechol derivatives, both native
and modified enzymes were analysed by electrospray mass
spectrometry. In good agreement with previously published
work4) showing that inhibition of Fe-SOD by catechol 8 was
associated with the formation of a complex which could not
be dissociated by gel-filtration, this spectrometry experiment
spectrophotometer. Merck silica gel 60 (70—230 mesh) was used for co-
lumn chromatography. All chemicals were obtained from Aldrich, and used
without further purification. Superoxide dismutases from bovine erythro-
cytes and from E. coli, and tyrosinase from mushroom were purchased from
Sigma.
Catechol Synthesis. tert-Butyl N-{2-[2,3-Dihydroxybenzoyl)amino]-
ethyl}carbamate 2 To a solution of 2,3-dihydroxybenzoic acid (0.96 g,
clearly demonstrated the irreversible modification of Fe- 6.2 mmol) and [(tert-butyloxycarbonyl)amino]ethylamine (1 g, 6.2 mmol) in

May 2002
581
(NH3) m/z: 333 [MϩHϩ], 350 [MϩNHϩ4 ].
dry tetrahydrofuran (THF, 8 ml) was added dropwise at 0 °C a solution of
1,3-dicyclohexylcarbodiimide (1.42 g, 6.9 mmol) in THF (3 ml). The solu-
tion was stirred at room temperature for 18 h. The reaction mixture was fil-
tered, and the filtrate evaporated to dryness. The residue was treated with
ethyl acetate and then washed with 5% aq. sodium hydrogencarbonate. The
organic layer was dried over magnesium sulfate, and the solvent evaporated
under reduced pressure. The residue was column chromatographed using a
mixture of dichloromethane and ethyl acetate (9 : 1) to yield 0.66 g of 2
(yield: 66 %). 1H-NMR (CDCl3) d: 1.43 (9H, s), 3.49 (4H, m), 5.09 (1H, m),
5.95 (1H, s), 6.72 (1H, t, Jϭ8.0 Hz), 7.02 (2H, d, Jϭ8.0 Hz). 13C-NMR
(CDCl3) d: 28.3, 39.5, 42.4, 80.5, 114.0, 116.7, 118.0, 118.6, 145.7, 149.1,
158.0, 170.6. IR (KBr) cmϪ1: 3356, 1695, 1637. Desorption chemical ion-
ization (DCI)-MS (NH3) m/z: 297 [MϩHϩ]; 314 [MϩNHϩ4 ].
N1-{6-[2,3-Dihydroxybenzoyl)amino]hexyl}-2,3-dihydroxybenzamide 8
To a solution of 1,6-hexanediamine (0.069 g, 0.6 mmol) in THF (5 ml), was
added dropwise at 0 °C a solution of 6 (0.3 g, 1.12 mmol) in THF (5 ml). The
reaction mixture was stirred at room temperature for 18 h, and the solvent
evaporated. The residue was chromatographed on silica gel using a mixture
of dichloromethane/methanol (9 : 1) to give 0.08 g of 8 (yield: 35%). 1H-
NMR (DMSO-d6) d: 1.20—1.34 and 3.24—3.31 (12H, m), 6.67 (2H, t,
Jϭ8.0 Hz), 6.90 and 7.28 (4H, d, Jϭ8.0 Hz), 8.77 (1H, s). 13C-NMR
(DMSO-d6) d: 26.1, 26.9, 28.7, 114.8, 117.0, 117.7, 118.6, 146.1, 149.6,
169.6. IR (KBr) cmϪ1: 1638. DCI-MS (NH3) m/z: 389 [MϩHϩ], 406
[MϩNHϩ4 ].
Inhibition Studies Superoxide dismutase activities were determined by
UV–visible spectroscopy at pH 8.2 by measuring the rate of the auto-oxida-
tion of pyrogallol (4—5 mM) at 420 nm. Enzymatic inhibition studies were
carried out as follows: a solution of inhibitor (10 ml) in dimethylsulfoxide (1
to 100 mM) was added to a solution (990 ml) of Fe-SOD (0.30mM) or Cu/Zn-
SOD (0.34 mM) in 50 mM Tris buffer (1 mM EDTA, pH adjusted to 8.2 with
cacodylic acid). The percentage of inhibition was calculated as follows: I
%ϭ[VI-SODϪVSOD]/[VIϪVSOD]ϫ100, where VI is the rate of auto-oxidation of
pyrogallol in the presence of inhibitor, VI-SOD in the presence of SOD and in-
hibitor, and VSOD in the presence of SOD.
To determine the dose-dependence inhibition by catechol compounds,
various concentrations were incubated with Fe-SOD or Cu/Zn-SOD (or
0.34 mM) for 24 h at 25 °C. The time-dependence was determined by periodi-
cal assays of incubation with or without catechol compounds at various con-
centrations with Fe-SOD.
Catechols Oxidation Oxidation reactions of compound 5 (400 mM) in
the absence of the enzyme, or in the presence of Cu/Zn-SOD (0.34 mM), Fe-
SOD (0.30 mM) or tyrosinase (5 U/ml) were monitored by UV–visible spec-
trophotometry at 280 nm in a 50 mM Tris buffer pH 8.2.
Electrospray Mass Spectrometry Compound 5 (40 mM) was dissolved
in a 100 mM ammonium hydrogenocarbonate buffer pH 8.0 containing Fe-
SOD (30 mM) at 25 °C. After 24 h, the solution (25 ml) was then diluted in
25 ml of a mixture of acetonitrile and water (1/1) containing 0.1% formic
acid. The solution was directly injected into the electrospray interface of a
LC MS/MS mass spectrometer (API 365, Perkin Elmer SCIEX). Spectra
were scanned over the range of 500—1600 m/z.
N1-(2-Ammonioethyl)-2,3-dihydroxybenzamide Chloride 3 A solu-
tion of 2 (0.3 g, 1 mmol) in dichloromethane (5 ml) and trifluoroacetic acid
(5 ml) was stirred for 16 h at room temperature. The reaction mixture was
then evaporated to dryness under reduced pressure, and several co-evapora-
tions with water were carried out. The residue was dissolved in ethanol
(5 ml), and 12 M hydrochloric acid (84 ml) was added. The solution was fil-
1
tered to give 0.17 g of 3 (yield: 72%). H-NMR (DMSO-d6) d: 3.01—3.02
and 3.56—3.58 (4H, m), 6.68 (1H, t, Jϭ8.0 Hz), 6.96 and 7.43 (2H, d,
Jϭ8.0 Hz), 8.23 (3H, s), 9.13 (1H, s). 13C-NMR (DMSO-d6) d: 36.7, 38.2,
114.8, 117.6, 117.8, 118.9, 146.1, 149.5, 170.2. IR (KBr) cmϪ1: 1640. DCI-
MS (NH3) m/z: 197 [MϩHϩ].
Ethyl 2-[(2,3-Dihydroxybenzoyl)amino]acetate 4 To a solution of
glycine ethyl ester hydrochloride (0.905 g, 6.48 mmol), triethylamine (0.657
g, 6.49 mmol) and 2,3-dihydroxybenzoic acid (1 g, 6.49 mmol) in dry THF
(8 ml) was added dropwise at 0 °C 1,3-dicyclohexylcarbodiimide (1.47 g,
7.14 mmol) in dry THF (3 ml). The solution was stirred at room temperature
for 16 h. The reaction mixture was filtered and the filtrate evaporated to dry-
ness. The residue was diluted with ethyl acetate, washed with 0.01 M aq. hy-
drochloric acid and water. The organic layer was dried over magnesium sul-
fate and evaporated to dryness. The product was chromatographed using
dichloromethane as eluent to give 0.62 g of 4 as a white powder (yield:
1
40%). H-NMR (DMSO-d6) d: 1.21 (3H, t, Jϭ7.0 Hz), 4.04 (2H, d, Jϭ6.0
Hz), 4.17 (2H, q, Jϭ7.0 Hz), 6.72 (1H, t, Jϭ8.0 Hz), 6.94 (1H, dd, Jϭ8.0,
1.0 Hz), 7.30 (1H, dd, Jϭ8.0, 1.0 Hz), 9.17 (1H, t, Jϭ7.0 Hz). 13C-NMR
(DMSO-d6) d: 14.0, 40.0, 80.5, 114.8, 117.5, 118.2, 119.0, 146.1, 149.2,
169.4, 169.8. IR (KBr) cmϪ1: 1723, 1655. DCI-MS: (NH3) m/z: 240
[MϩHϩ], 257 [MϩNHϩ4 ].
Acknowledgments We thank Catherine Claparols for excellent techni-
cal assistance in the mass spectrometry field. Laurent Soulère is the recipient
of a grant from the Ministère de la Recherche (France).
2-[(2,3-Dihydroxybenzoyl)amino]acetic Acid 5 Ester 4 (0.5 g, 2.1
mmol) was dissolved in 1 M aq. sodium hydroxide (25 ml) and the solution
was stirred at room temperature for 4 h. After filtration, the filtrate was acidi-
fied with diluted sulfuric acid, and the residue was extracted with ethyl ace-
tate. The organic layer was dried over magnesium sulfate and evaporated to
dryness to yield 0.37 g of 5 as a brown powder (yield: 84%). 1H-NMR
(DMSO-d6) d: 3.97 (2H, d, Jϭ6.0 Hz), 6.71 (1H, t, Jϭ8.0 Hz), 6.93 (1H, dd,
Jϭ8.0, 1.0 Hz), 7.31 (1H, dd, Jϭ8.0, 1.0 Hz), 9.12 (1H, t, Jϭ6.0 Hz), 12.31
(1H, s). 13C-NMR (DMSO-d6) d: 40.9, 114.8, 117.4, 118.1, 118.9, 146.1,
149.2, 169.7, 170.8. IR (KBr) cmϪ1: 1731, 1650. DCI-MS (NH3) m/z: 212
[MϩHϩ], 224 [MϩNHϩ4 ].
References
1) Fridovich I., Annu. Rev. Biochem., 64, 97—112 (1995).
2) Parker M. W., Blake C. C., FEBS Lett., 229, 377—382 (1988).
3) Meshnick S. R., Eaton J. W., Biochem. Biophys. Res. Commun., 102,
970—976 (1981).
4) Meshnick S. R., Kitchener K. R., Trang N. L., Biochem. Pharmacol.,
34, 3147—3152 (1985).
5) Iverson S. L., Shen L., Anlar N., Bolton J. L., Chem. Res. Toxicol., 9,
492—499 (1996).
6) Gantchev T. G., Hunting D. J., Mol. Pharmacol., 53, 422—428 (1998).
7) Zhang F., Fan P. W., Liu X., Shen L., van Breeman R. B., Bolton J. L.,
Chem. Res. Toxicol., 13, 53—62 (2000).
2,5-Dioxotetrahydro-1H-1-pyrrolyl 2,3-Dihydroxybenzoate 6 To a so-
lution of N-hydroxysuccinimide (1.495 g, 13 mmol) and 2,3-dihydroxyben-
zoic acid (2 g, 13 mmol) in dry THF (16 ml) was added dropwise at 0 °C a
solution of 1-3-dicyclohexylcarbodiimide (3.21 g, 15.6 mmol) in dry THF
(6 ml). The solution was stirred at room temperature for 16 h. The reaction
mixture was filtered, and the filtrate evaporated to dryness. The residue was
taken up with ethanol, and the insoluble material was filtered to give 1.26 g
8) Mayor M., Scheffold R., Walder L., Helv. Chim. Acta, 80, 1183—1189
(1997).
9) Marklund S., Marklund G., Eur. J. Biochem., 47, 469—474 (1974).
10) Misra H. P., Fridovich I., J. Biol. Chem., 247, 3170—3175 (1972).
11) Heikkila R. E., Cabbat F., Anal. Biochem., 75, 356—362 (1976).
12) Ishii T., Iwahashi H., Sugata R., Kido R., Fridovich I., Arch. Biochem.
Biophys., 276, 248—250 (1990).
13) Ishii T., Fridovich I., Free Radic. Biol. Med., 8, 21—24 (1990).
14) Munday R., Free Radic. Biol. Med., 26, 1475—1479 (1999).
15) Munday R., Free Radic. Res., 32, 245—253 (2000).
16) Chang M., Shin Y. G., van Breemen R. B., Blond S. Y., Bolton J. L.,
Biochemistry, 40, 4811—4820 (2001).
1
of pure 6 (yield: 36%). H-NMR (DMSO-d6) d: 2.87 (4H, s), 6.81 (1H, t,
Jϭ8 Hz), 7.13 (1H, dd, Jϭ8.0, 1.0 Hz), 7.30 (1H, dd, Jϭ8.0, 1.0 Hz). 13C-
NMR (DMSO-d6) d: 25.5, 110.9, 119.1, 120.6, 120.8, 146.4, 148.8, 161.5,
170.4. IR (KBr) cmϪ1: 1729. DCI-MS (NH3) m/z: 269 [MϩNHϩ4 ], 286
[MϩN2Hϩ7 ].
N1-{2-[2,3-Dihydroxybenzoyl)amino]ethyl}-2,3-dihydroxybenzamide 7
To a solution of ethylenediamine (0.06 g, 0.99 mmol) in THF (5 ml) was
added dropwise at 0 °C a solution of 6 (0.5 g, 1.98 mmol) in THF (5 ml). The
solution was stirred overnight at room temperature. The solvent was evapo-
rated to dryness and the residue was taken up with ethyl acetate and washed
with 0.01 M hydrochloric acid. The organic layer was dried over magnesium
sulfate and evaporated to dryness. Recrystallisation from dichloromethane
17) Lah M. S., Dixon M. M., Pattridge K. A., Stallings W. C., Fee J. A.,
Ludwig M. L., Biochemistry, 34, 1646—1660 (1995).
18) Sines J. J., Allison S. A., McCammon J. A., Biochemistry, 29, 9403—
9412 (1990).
1
gave 0.135 g of 7 as a white powder (yield: 41%). H-NMR (DMSO-d6) d:
3.49 (4H, s), 6.68 (2H, t, Jϭ8.0 Hz), 6.92 (2H, dd, Jϭ8.0, 1.0 Hz), 7.27 (2H,
dd, Jϭ8.0, 1.0 Hz), 8.93 (1H, s). 13C-NMR (DMSO-d6) d: 38.5, 114.9,
117.1, 117.8, 118.8, 146.1, 149.6, 170.1. IR (KBr) cmϪ1: 1644. DCI-MS
19) Sines J., Allison S., Wierzbicki A., McCammon J. A., J. Phys. Chem.,
94, 959—961 (1990).
20) Getzoff E. D., Cabelli D. E., Fisher C. L., Parge H. E., Viezzoli M. S.,

582
Vol. 50, No. 5
Banci L., Hallewell R. A., Nature (London), 358, 347—351 (1992).
Sci., 7, 2354—2358 (1998).
21) Polticelli F., Bottaro G., Battistoni A., Carri M. T., Djinovic-Carugo 23) Borders C. L., Jr., Bjerrum M. J., Schirmer M. A., Oliver S. G., Bio-
K., Bolognesi M., O’Neill P., Rotilio G., Desideri A., Biochemistry,
34, 6043—6049 (1995).
22) Polticelli F., Battistoni A., O’Neill P., Rotilio G., Desideri A., Protein
chemistry, 37, 11323—11331 (1998).
24) Chan V. W., Bjerrum M. J., Borders C. L., Jr., Arch. Biochem.
Biophys., 279, 195—201 (1990).