Chem.-Biol. Interactions, 85 (1992) 35-48 35 Elsevier Scientific Publishers Ireland Ltd.

EVIDENCE FOR FREE RADICAL FORMATION DURING HORSERADISH PEROXIDASE-CATALYZED N-DEMETHYLATION OF CRYSTAL VIOLET

FERNANDA R. GADELHA a, PHILLIP M. HANNA b, RONALD P. MASON b and ROBERTO DOCAMPO a

aDepartment of Veterinary Pathobiology, University of Illinois, Urbana, Illinois 61801 and bLaboratory of Molecular Biophysics, National Institute of Environmental Health Sciences, National Institutes of Health, P.O. Box 12233, Research Triangle Park, North Carolina 27709 (USA)

(Received May 18th, 1992) (Revision received August 28th, 1992) (Accepted September 1st, 1992)

SUMMARY

Crystal violet (gentian violet) can undergo an oxidative metabolism, catalyzed by horseradish peroxidase, resulting in formaldehyde formation. The N- demethylation reaction was strongly inhibited by reduced glutathione. Evidence for the formation of a crystal violet radical during the horseradish peroxidase catalyzed reaction was the detection of thiyl and ascorbate radicals from glutathione and ascorbate, respectively. The concentration of radicals from both compounds was significantly increased in the presence of crystal violet. Oxygen uptake was stimulated when glutathione was present in the system and this oxy- gen uptake was dependent on the dye and enzyme concentration. Oxygen uptake did not occur when ascorbate, instead of glutathione, was present in the system. However, when glutathione was present, ascorbate totally inhibited the glutathione-stimulated oxygen uptake in the crystal violet/horseradish perox- idase/hydrogen peroxide system. Although a weak ESR spectrum from a crystal violet-derived free radical was detected when the dye reacted with H202 and horseradish peroxidase, using the fast flow technique, this spectrum could not be interpreted.

Key words: Crystal violet -- Horseradish peroxidase -- Thiyl radical -- Ascorbate radical -- N-demethylation -- Superoxide anion

Correspondence to: Roberto Docampo, Department of Veterinary Pathobiology, University of Illinois, 2001 South Lincoln Avenue, Urbana, IL 61801, USA. Abbreviations: Diethylenetriaminepentaacetic acid, DTPA; 5,5-dimethyl-l-pyrroline N-oxide, DMPO; horseradish peroxidase, HRP; reduced glutathione, GSH; oxidized glutathione, GSSG; 2- methyl-2-nitrosopropane, t-NB.

0009-2797/92/$05.00 © 1992 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

36

'CH3,2N

+N(CH3)2CI-

C ~N(CH3)2 Fig. 1. Hexamethylpararosaniline chloride (crystal violet, gentian violet).

INTRODUCTION

Triphenylmethane dyes are widely used as biological stains, in dyes for wood, silk and food and in cosmetics [1]. The irreversible fixation of crystal violet (gentian violet, N~V2¢' ~ ' ~V" ~V "- hexamethylpararosaniline, Fig. 1) by Gram- positive bacteria is the basis of the Gram stain [2]. Crystal violet has also been used as a mycostatic agent in poultry feed and for the control of fungal and intestinal parasites in humans because it has long been known to inhibit the growth of certain fungi, bacteria and parasites [1]. Since Nussenzweig et al. [3] demonstrated the in vitro activity of crystal violet against bloodstream trypomastigotes of Trypanosoma cruzi, this compound has been widely used in blood banks in Latin America in attempts to eliminate blood transmission of Chagas' disease [4].

In addition to a reductive metabolism with the formation of a carbon-centered free radical [5] and its leucoderivative [6], crystal violet can be oxidatively N- demethylated by liver microsomes of different animals [7] leading to the forma- tion of N,N,N',N',N "-pentamethylpararosaniline and the isomeric N,N,N',N'- and N,N,N',N"-tetramethylpararosaniline. It can also be N- demethylated by lignin peroxidase leading to the formation ofN, N,N' ,N' ,N "-p- entamethylpararosaniline, N,N,N' ,N'-tetramethylpararosaniline and N,N,N'- trimethylpararosaniline [8]. N-demethylation reactions catalyzed by peroxidases such as horseradish peroxidase (HRP) [9] involve the formation of free radical intermediates and other reactive species that are able to initiate oxidative stress or bind to macromolecules [10, 11]. The mechanism of the N-demethylation of N- substituted aromatic amines like crystal violet is not well understood, but it ap- pears that one-electron oxidation leads to the formation of nitrogen-centered cat- ion radicals [11]. The subsequent mechanism of N-dealkylation, however, is dependent upon the structure of the nitrogen-centered radical formed [11]. The oxidation of crystal violet by peroxidases may have toxicological significance since the completely demethylated crystal violet derivative, leucopararosaniline,

37

has been demonstrated to be carcinogenic in rats [12]. In this regard, two recent reports [13,14] have indicated that crystal violet is a carcinogen in rodents at several organ sites.

Glutathione (GSH) is known to reduce the free radical metabolites resulting from the dealkylation of aminopyrine [15] and p-phenetidine [16], leading to the formation of the unchanged parent molecule and the glutathionyl radical and radicals derived therefrom. This type of reaction is known as thiol pumping and it has been proposed as an important mechanism for the detoxification of free radical metabolites [17]. In the presence of oxygen, thiol pumping leads to the formation of superoxide anion with concomitant oxygen consumption [18]. Free radicals can also be reduced by other radical scavengers like ascorbic acid [19,20].

In this paper, we report that crystal violet can undergo an N-demethylation reaction catalyzed by horseradish peroxidase, resulting in the formation of free radical derivatives. In addition, we provide evidence of thiol pumping in the presence of glutathione and the effective oxidation of ascorbate by the crystal violet-derived free radical.

MATERIALS AND METHODS

Crystal violet, glutathione (GSH), horseradish peroxidase (HRP), (type VI, EC 1.11.1.7), hydrogen peroxide, sodium ascorbate, 5,5-dimethyl-l-pyrroline N-

CH 3 I

R--N- -CH 3

e -

CH a I

R--+N--CH 3

CH a I

R - - N = C H 2

+ H20

CH3 I

R--NH + ,o

4-

CH20

CH 3 I

R--N- -CH 3

Scheme 1. Mechanisms possible for the oxidation of crystal violet. (R = substituted aromatic ring.)

38

oxide (DMPO), 2-methyl-2-nitrosopropane (t-NB) and diethylenetriaminepen- taacetic acid (DTPA) were purchased from Sigma. Catalase was from Boehr- inger Mannheim. DMPO was distilled prior to use. All other reagents were of analytical grade.

Electron spin resonance spectra were recorded at room temperature, 24°C, using a Bruker ER 200D spectrometer operating at 9.77 GHz, with a 100-kHz modulation frequency. A microwave power of 20 mW and a modulation amplitude of 0.25 G were employed. The reaction was started with the addition of horseradish peroxidase, vortex-mixed and then aspirated to a quartz aqueous cell (Wilmad, Buena, N J) prepositioned in a TM110 microwave cavity. The scan was initiated 6 s or 60 s after mixing, as indicated. Spectra were recorded on an IBM-compatible computer interfaced to the spectrometer. The solutions of glutathione and ascorbate were prepared immediately prior to use and kept under nitrogen. Flow experiments were conducted with a quartz fast-flow mix- ing chamber flat cell obtained from Wilmad Glass Co., Buena, NJ (type WG-Q4, modified flat cell, 17 mm width). Reagents were prepared in two 2-1 bottles. Outlets at the bases of the bottles were connected to the inlets of the flat cell with Tygon tubing. Gravity flow from a height of 2 m was regulated by Gilmont compact flow meters [17].

Oxygen consumption was measured using a Gilson oxygraph with a Clark-type electrode. All experiments were performed at room temperature and the incuba- tion conditions were as described in the figure legends.

The N-demethylation activity of horseradish peroxidase was determined by

3O

i ..~= 10

0 1 0 0 2 0 0 3 0 0

Time (sec)

Fig. 2. Time course showing the effect of GSH on the N-demethylation of crystal violet catalyzed by HRP. The reaction mixture contained 25 ~M crystal violet, with (0) or without (O) 1 mM GSH, 16 ~M hydrogen peroxide, 25 ~g/ml HRP and 0.1 M Tris-HCl buffer, pH 7.4, in a final volume of 0.5 ml. The reaction was initiated by the addition of HRP, incubated at room temperature for the times indicated and terminated by the addition of 7.5% trichloroacetic acid to the reaction mixture. Formaldehyde was assayed by the Nash procedure (see Ref. 21).

39

measuring the amount of formaldehyde formed using the Nash spectrophoto- metric assay [21]. Values represent the averages of at least three experiments. Light had no effect on these experiments.

RESULTS

Incubation of crystal violet with HRP and H202 resulted in the formation of formaldehyde. The amount of formaldehyde formed increased with HRP concen- tration until at least 100 ~g/ml (data not shown). The time course for the HRP- catalyzed formation of formaldehyde from crystal violet is shown in Fig. 2. Under the conditions used, the rate of formaldehyde formation was linear with time for the first 2 min of reaction. Figure 2 also indicates that more than one methyl group of crystal violet is lost during this reaction. Thus, at 300 s about 25 nmol of formaldehyde were formed in a 0.5-ml incubation volume (50 ~M formaldehyde) when 25 ~M crystal violet was present (Fig. 2). This is in agree- ment with the results reported for the N-demethylation of crystal violet by liver microsomes of different animals [7] or by lignin peroxidase [8], where the major products of crystal violet found were the pentamethyl, tetramethyl [7,8] and trimethylpararosaniline derivatives [8].

Since N-demethylation reactions of N-substituted aromatic amines catalyzed by peroxidases involve the formation of nitrogen-centered cation radicals [11], we tried to detect a crystal violet radical derivative (CV" +) by either direct ESR at low pH (pH 4 - 5), where the CV" + may be more stable, or by using the ESR fast-flow technique. Although a signal was detected upon incubation of crystal violet, H202 and horseradish peroxidase using the ESR fast-flow tech- nique (not shown), this was very weak and not interpretable.

TABLE I

EFFECT OF ASCORBATE AND GSH ON DEMETHYLATION OF CRYSTAL VIOLET

THE RATE OF HRP-CATALYZED N-

Crystal violet (~M) % Inhibition

+ GSH + ascorbate

10 100.0 100.0 25 87.5 ± 2.5 100.0 50 77.8 ± 4.8 93.7 ± 1.1

100 30.7 ± 5.8 43.6 ± 7.7

The reaction mixture contained the indicated amounts of crystal violet, 25 ~g/ml HRP, 16 ~M hydrogen peroxide, with 1 mM GSH or 1 mM ascorbate and 0.1 M Tris HCl buffer, pH 7.4, in a final volume of 0.5 ml, The reaction was initiated with HRP, incubated at room temperature for 1 rain and terminated by the addition of 7.5% trichloroacetic acid. Formaldehyde formation was assayed by the Nash procedure (see Ref. 21). Formaldehyde formation was 6.7 ± 1.5, 15.3 ± 4.0, 28.3 ± 5.5 and 107.7 ± 15.4 nmol formaldehyde per min in the presence of 10, 25, 50 and 100 ~M crystal violet, respectively.

40

Some authors [9] have suggested that N-demethylation of some N-substituted aromatic amines such as N~'-dimethylaniline by horseradish peroxidase in- volves hydrogen atom abstraction from the methyl group, leading directly to the formation of a neutral carbon-centered radical rather than to the formation of a nitrogen-centered radical cation. Carbon-centered radicals are easily detected by spin-trapping techniques [11]. However, incubation of crystal violet with HzO~ and horseradish peroxidase in the presence of either DMPO (Fig. 5C) or t-NB (not shown) did not result in the detection of a carbon-centered radical. These experiments were also performed using a wide range of substrate and enzyme concentrations at pH 4.0 (not shown). These results indicate that crystal violet is probably oxidized by horseradish peroxidase to an aromatic nitrogen- centered radical, which is not reactive with spin traps, with little or no formation of a carbon-centered radical.

Since GSH is known to reduce the free radical metabolites resulting from the N-demethylation of aromatic compounds [15,16], leading to the formation of the

II

o

H202

~ " C

B

A

H202

2 rnin

G F

E

Fig. 3. Oxygen uptake during the oxidation of crystal violet by HRP. (A) Oxygen consumption was determined in a medium containing of 25 #M crystal violet, 25 gg/ml HRP, 1 mM GSH, 16 gM hydrogen peroxide and 0.1 M Tris- HC1 buffer, pH 7.4, in a final volume of 1.7 ml. The reaction was initiated by the addition of hydrogen peroxide. (B) As in A, but in the absence of crystal violet. (C) As in A, but in the absence of GSH. (D) As in A, but in the presence of 1 mM ascorbate. (E) As in A, but in the presence of ascorbate and absence of GSH. (F) As in A, but in the presence of ascorbate and absence of crystal violet. (G) As in A, but in the presence of ascorbate and absence of crystal violet and GSH. (E and F, are superimposed.)

A B

unchanged parent molecule and the glutathionyl radical and radicals derived therefrom (thiol pumping), we investigated the effect of GSH and other reducing agents such as ascorbate on the HRP/H202-catalyzed N-demethylation of crystal violet. As shown in Fig. 2 and Table I, addition of GSH (Fig. 2, Table I) or ascorbate (Table I) caused a marked inhibition of this reaction, ascorbate being more efficient than GSH (Table I). The effects of glutathione and ascorbate on crystal violet metabolism by horseradish peroxidase were then further in- vestigated by following the oxygen uptake of the system. With the addition of HRP to an aqueous solution of I mM GSH, a small amount of oxygen uptake was observed as a result of direct oxidation of GSH by the enzyme and the reaction of the resulting radicals with molecular oxygen (Fig. 3B) [18]. Upon addition of crystal violet, oxygen uptake was significantly increased (Fig. 3A). Ascorbate at a concentration of 1 mM completely inhibited oxygen consumption (Fig. 3D). There was no oxygen consumption in the absence of GSH (Fig. 3C). Upon the addition of ascorbate to the system without GSH (Fig. 3E), or without crystal violet (Fig. 3F), or with ascorbate alone (Fig. 3G), no oxygen uptake was observed.

Figure 4 shows that the amount of oxygen consumed when crystal violet was oxidized by HRP and H202 in the presence of GSH increased with the enzyme

300'

200'

100"

0 0

• , , , , ~ . , " 3 , . , . ' ~

200 4 0 0 6 0 0 800 1 0 0 0 50 100 150 200 2 5 0

o E , m

E , -

e- > , o x o o c

O , -

U~ O~

O X o o c (n

7 0

60"

5 0

4 0

3 0

2 0

1 0

0 0

41

H R P (pM) C r y s t a l v i o l e t (I~M)

Fig. 4. Oxygen uptake during the oxidation of crystal violet by HRP as a function of HRP (h) and crystal violet (B) concentration. (A) The reaction mixture contained the concentrations of HRP indicated, 16 ~M hydrogen peroxide, 1 mM glutathione, in the presence (e) or absence (O) of 25 ~M crystal violet and 0.1 M Tris-HC1 buffer, pH 7.4, in a final volume of 1.7 ml. The reaction was initiated by the addition of hydrogen peroxide. (B) The reaction mixture contained 25 ~g/ml HRP, 16 ~M hydrogen peroxide, the concentrations of crystal violet indicated, in the presence (e) or absence (O) of 1 mM glutathione and 0.1 M Tris-HC1 buffer, pH 7.4, in a final volume of 1.7 ml. The reaction was initiated by the addition of hydrogen peroxide.

42

(Fig. 4A) and the crystal violet (Fig. 4B) concentration. At higher enzyme con- centration (> 250 t~g/ml) (Fig. 4B), horseradish peroxidase catalyzed a direct one- electron oxidation of GSH, as reported by Harman et al. [18]. Oxygen consump- tion from the oxidation of crystal violet was not significant. The lack of oxygen consumption when crystal violet alone was present (Figs. 3 C and 4B) confirms that crystal violet is not oxidized to a carbon-centered free radical derivative.

J I Complete

- Crystal

C

D - H R P

E - H 2 0 2

20 Gauss

Fig. 5. ESR spectra of the DMPO/glutathione thiyl radical adduct generated in a system of crystal violet and HRP. Data acquisition was initiated 60 s after starting the reaction. Instrumental condi- tions were: 20 mW microwave power, 0.25 G modulation amplitude, 164 ms time constant and 40 G/rain scan rate. (A) Incubation containing 25 t~M crystal violet, 1 mM glutathione, 100 mM DMPO, 25 ~g/ml HRP, 16 ~M hydrogen peroxide and 0.1 M Tris-HC1 buffer, pH 7.4. (B) As in A, but in the absence of crystal violet. (C) As in A, but in the absence of glutathione. (D) As in A, but in the absence of HRP. (E) As in A, but in the absence of hydrogen peroxide.

43

Carbon-centered radicals would most likely react with oxygen to form peroxyl radicals [11|.

In order to detect the free radicals that were formed during these reactions, spin-trapping experiments were performed. A four-line ESR signal correspon- ding to the DMPO/glutathione thiyl radical adduct was obtained with the com-

/ ~ Complete

SH

C ~ ~ ~ - Crystal ~ o l e t

D - HRP

E

F

- H202

- H202 + Catalase

G - Ascorbate

1 Gauss t

Fig. 6. ESR spectra of the aseorbate radical generated in a system of crystal violet and HRP. Data acquisition was initiated 6 s after starting the reaction. Instrumental conditions were: 20 mW microwave power, 0.25 G modulation amplitude, 164 ms time constant and 30 G/rain scan rate. (A) Incubation containing 25 t~M crystal violet, 1 mM aseorbate, 2.5 ag/ml HRP, 16 t~M hydrogen perox- ide, 0.5 mM DTPA and 0.1 M Tris-HC1 buffer, pH 7.4. (B) As in A, but in the presence of I mM GSH. (C) As in A, but in the absence of crystal violet. (D) As in A, but in the absence of HRP. (E) As in A, but in the absence of hydrogen peroxide. (F) As in A, but in the absence of HzO z and presence of 50 ~g/ml of eatalase. (G) As in A, but in the absence of aseorbate.

44

plete system, i.e. crystal violet/GSH/HRP/H202/DMPO (Fig. 5A). The adduct has a distinctive spectrum (a N -- 15.1 G, aB H = 16.2 G) with.hyperfine split- tings similar to those reported by H a m a n et al [18]. Without crystal violet, the same signal, but with lower amplitude, was observed due to the direct oxidation of GSH by the enzyme system (Fig. 5B) [18]. In the absence of GSH, the oxidized product of DMPO, referred to as DMPOX, was detected (Fig. 5C). The hyperfine coupling c o n s t a n t s (a N -- 7.2 G, a ~ n -- 4.2 G) are similar to those reported by Floyd and Soong [22]. In this regard, the very stable DMPOX has been observed previously as a product of horseradish peroxidase-hydrogen peroxide oxidation reactions [23]. The crystal violet radical (CV" ÷) could not be detected under these conditions, supporting the idea that the radical decays very rapidly. No signal was detected in the absence of HRP (Fig. 5D) or H202 (Fig. 5E).

ESR experiments were done to detect ascorbate radical formation by the crystal violet/HRP/H202 system (Fig. 6). The presence of ascorbate in the crystal violet/HRP/H202 system led to the formation of the ascorbate radical (Fig. 6A). The ascorbate radical concentration was about 25% lower in the presence of GSH (Fig. 6B) and about 50% lower in the absence of crystal violet (Fig. 6C). Weak signals were detected in the absence of HRP (Fig. 6D) or H202 (Fig. 6E). A weak signal in the absence of H202 (Fig. 6E) was almost completely abolished in the presence of 50 ~g/ml of catalase (Fig. 6F), thus indicating the presence of some H202 in the crystal violet solutions, probably arising from the photochemical activation of the dye [24]. The signal shown in Fig. 6F could be due to an adventitious metal-catalyzed formation of ascorbyl radical. No signal was detected in the absence of ascorbate (Fig. 6G).

DISCUSSION

Incubation of crystal violet with HRP and hydrogen peroxide resulted in the formation of formaldehyde. Scheme 1 shows some possible mechanisms for the horseradish peroxidase-catalyzed N-demethylation of crystal violet. One possible mechanism involves an initial one-electron oxidation ( - e -) of the amine to form a nitrogen-centered radical (CV" +), which loses a hydrogen atom ( - H ' ) to form an iminium cation. The iminium cation is hydrolyzed to the monomethyl amine derivative and formaldehyde. Another possible mechanism is the oxida- tion to a nitrogen-centered radical, which then disproportionates to the same im- inium cation and crystal violet. Other possible mechanisms involving the formation of carbon-centered radicals such as those described for the N- demethylation of N-CH2-COOH-substituted amines [11] can be excluded since we were unable to detect any carbon-centered radicals using spin trapping agents and no oxygen consumption was detected during the HRP/H202- catalyzed N-demethylation of crystal violet.

Formaldehyde formation was inhibited by GSH and ascorbate. In addition, GSH stimulated the oxygen uptake of the crystal violet/HRP/H202 system. The rate of oxygen uptake was dependent on the HRP (Fig. 4A) and dye concentra- tion (Fig. 4B). GSH oxidation to thiyl radical by HRP (Fig. 5B) was greatly stimulated by crystal violet (Fig. 5A). The stimulation of the formation of this

45

radical was dependent upon the presence of crystal violet, HRP and hydrogen peroxide. These observations indicate that a free radical from crystal violet was formed which was able to react with GSH to form the thiyl radical. The crystal violet radical was reduced back to the parent compound in the presence of GSH and N-demethylation was inhibited, similar to the mechanism proposed for aminopyrine peroxidase-catalyzed N-demethylation [25]. The concomitant stimulation of oxygen consumption in the presence of GSH suggests that the thiyl radical (GS ' ) formed in this system reacts with the glutathione anion (GS-) to generate the glutathione disulfide anion radical (GSSG)"-, which in turn can react with oxygen to generate superoxide anion [17].

A mechanism for the oxygen activation can be described by the following reactions:

HRP 2 CV + H202 + 2H ÷ - 2CV "÷ + 2 H20 (1)

In the presence of GSH:

CV "÷ + G S H - C V + GS" + H ÷ (2)

The thiyl radicals formed can react with GS- to form disulfide radical anions, which react with oxygen to form superoxide [17]:

GS" + GS- - (GSSG)" - (3)

(GSSG)'- + 02 L GSSG + O{ (4)

or with another molecule of GS" :

GS" + GS" - GSSG (5)

These chemical reactions, including superoxide anion formation (reaction 4) are well established (reviewed in Ref. 18). Although no superoxide anion could be detected using DMPO (Fig. 5), this could be attributed to its fast reaction with glutathione (k = 6.7 x 105 M -1 s -1) [26]:

O{ + GSH - GS" + H202 (6)

GS" can also be formed from the direct one-electron oxidation of GSH [18].

HRP 2 GSH + H202 = 2 GS" + 2H20 (7)

In addition, glutathione-derived peroxides could result from the known addi- tion reaction of thiyl radicals with oxygen [27]:

GS" + 02 - GSO0" (8)

46

Therefore, if thiyl radicals are formed in excessive amounts, oxidative stress could result. Oxidative stress could lead to modifications of protein thiols and is considered a major cause of cellular toxicity. Such a mechanism would be of tox- icological importance. In this regard, depression in the white blood cell count after intravenous administration of crystal violet has been reported [28], and a good correlation has been found between the incidence of agranulocitosis and drug activation by peroxidases [29].

In the presence of ascorbate, the thiyl radical could not be detected, thus in- dicating that ascorbate reacts more rapidly with the crystal violet radical than does GSH. When ascorbate is present, the reaction of CV" ÷ with GSH is prevented because of reaction 9:

CV" ÷ + AH- * CV + A ' - + H + (9)

A similar finding was also illustrated in the oxygen uptake experiments (Fig. 3); oxygen consumption was rapid in the presence of GSH. Upon addition of ascorbate, no oxygen uptake was detected since the ascorbate semidione free radical reacts with oxygen very slowly [30] (Fig. 3D).

As a reducing agent and radical scavenger, ascorbic acid was able to reduce the crystal violet radical to the parent compound and subsequently inhibit its N- demethylation reaction. As illustrated in Fig. 6, when ascorbate was present in the crystal violet/HRP/H2OJGSH system, only the ascorbate radical could be detected. Therefore, ascorbate totally inhibited the reaction of the crystal violet radical with GSH in this system. Ascorbate oxidation to the ascorbate radical by HRP (Fig. 6C) was stimulated by crystal violet (Fig. 6A). This stimulation was slightly inhibited by GSH (Fig. 6B), implying that the crystal violet radical preferentially reacts with ascorbate.

Ascorbic acid is a physiological antioxidant of major importance for protection against diseases and degenerative processes caused by oxidative stress [31]. Our data suggest a protective role of ascorbate against toxicity related to crystal violet-derived radicals formed by the action of peroxidases. In this regard, crystal violet was shown to have a photodynamic action on T. cruzi trypo- mastigotes in the blood, via a reductive metabolism, that was enhanced by sodium ascorbate [32]. Since photoirradiation with visible light and addition of sodium ascorbate reduces significantly the effective dose and time of contact of crystal violet with T. cruzi-infected blood, addition of sodium ascorbate and photoirradiation has been proposed as a treatment of T. cruzi-infected blood in- tended for transfusion [32]. Our results indicate that the addition of sodium ascorbate to the blood, in addition to enhancing the photodynamic effect of crystal violet in the blood [32], would provide an additional protection against crystal violet-derived radicals formed through the action of mammalian peroxi- dases such as myeloperoxidase or prostaglandin synthase [10].

ACKNOWLEDGEMENTS

We thank Debora Weinert Moraes for help with preliminary experiments. This work was supported in part by NIH/NIAID grant AI-23259 and by a grant of the

47

United Nations/World Bank/World Health Organization Special Programme for Research and Training in Tropical Diseases to R.D.F.R.G. is a research fellow from the Conselho Nacional de Desenvolvimento Cientifico e TecnolSgico (CNPq, Brazil).

REFERENCES

1 R. Docampo and S.N.J. Moreno, The metabolism and mode of action of gentian violet, Drug Metab. Rev., 22 (1990) 161-178.

2 C. Gram, Uber die isolierte Farbung der Schizomyceten in Schnitt-und Trockenpraparaten, Fortschr. Med., 2 (1884) 185-190.

3 V. Nussenzweig, R. Sonntag, A. Biancalana, J.L. Pedreira de Freitas, V. Amato Neto and J. Kloetzel, Acao de corantes trifenil-met~nicos sobre o Trypanosoma cruzi "in vitro". Emprego de violeta de genciana na profilaxia da transmissao da molestia de Chagas por transfusao de sangue, Hospital (Rio de Janeiro), 44 (1953) 731-744.

4 R. Docampo, F.R. Gadelha and S.N.J. Moreno, Metabolism and mode of anti-Trypanosoma cruzi action of gentian violet, in: C.C.Wang (Ed.), Molecular and Immunological Aspects of Parasitism, American Association for the Advancement of Science, Washington DC, 1991, pp. 95-105.

5 W.G. Harrelson and R.P. Mason, Microsomal reduction of gentian violet. Evidence for cytochrome P-450-catalyzed free radical formation, Mol. Pharmacol., 22 (1982) 239-242.

6 J.J. McDonald and C.E. Cerniglia, Biotransformation of gentian violet to leucogentian violet by human, rat and chicken intestinal microflora, Drug Metab. Dispos., 12 (1984) 330-336.

7 J.J. McDonald, C.R. Breeden, B.M. North and R.W. Roth, Species and strain comparison of the metabolism of gentian violet by liver microsomes, J. Agric. Food Chem., 32 (1984) 596- 600.

8 J.A. Bumpus and B.J. Brock, Biodegradation of crystal violet by the white rot fungus Phanerochaete chrysosporium, Appl. Environ. Microbiol., 54 (1988) 1143-1150.

9 G.L. Kedderis and P.F. Hollenberg, Characterization of the N-demethylation reactions catalyz- ed by horseradish peroxidase, J. Biol. Chem., 258 (1983) 8129-8138.

10 R.S. Krauss and T.E. Eling, Arachidonic acid-dependent cooxidation. A potential pathway for the activation of chemical carcinogens in vivo (commentary), Biochem. Pharmacol., 33 (1984) 3319- 3324.

11 J. Van der Zee, D.R. Duling, R.P. Mason and T.E. Eling, The oxidation of N-substituted aromatic amines by horseradish peroxidase, J. Biol. Chem., 264 (1989) 19828-19836.

12 R.A.M. Case and J.T. Pearson, Tumours of the urinary bladder in workmen engaged in the manufacture and use of certain dyestuff intermediates in the British chemical industry, Br. J. Ind. Med., 11 (1954) 213-221.

13 N.A. Littlefield, B.-N. Blackwell, C.C. Hewitt and D.W. Gaylor, Chronic toxicity and carcino- genicity studies of gentian violet in mice, Fundam. Appl. Toxicol., 5 (1985),902-912.

14 N.A. Littlefield, D.W. Gaylor, B.-N. Blackwell and R.R. Allen, Chronic toxicity/carcinogenicity studies of gentian violet in Fischer 344 rats: two-generation exposure, Food Chem. Toxicol., 27 (1989) 239-247.

15 T.E. Eling, R.P. Mason and K. Sivarajah, The formation of aminopyrine cation radical by the peroxidase activity of prostaglandin H synthase and subsequent reactions of the radical, J. Biol. Chem., 260 {1985) 1601-1607.

16 D. Ross, K. Norbeck and P. Moldeus, The generation and subsequent fate of glutathionyl radicals in biological systems, J. Biol. Chem., 260 (1985) 15028-15032.

17 D.N.R. Rao, V. Fischer and R.P. Mason, Glutathione and ascorbate reduction of the acetaminophen radical formed by peroxidase, J. Biol. Chem., 265 (1990) 844- 847.

18 L.S. Harman, D.K. Carver, J. Schreiber and R.P. Mason, One-and two-electron oxidation of reduced glutathione by peroxidases, J. Biol. Chem., 261 (1986) 1642-1648.

19 T. Ohnishi, H. Yamazaki, T. Iyanagi, T. Nakamura and I. Yamazaki, One-electron-transfer reac- tions in biochemical systems, II. The reaction of free radicals formed in the enzymic oxidation, Biochim. Biophys. Acta, 172 (1969) 357-369.

48

20 M. Bando, H. Obazawa and T. Tanikawa, Scavenging of chlorpromazine cation radical by ascor- bic acid or giutathione, J. Free Rad. Biol. Med., 2 (1986) 261- 266.

21 T. Nash, The colorimetric estimation of formaldehyde by means of the Hantzsch reaction, Biochem. J., 55 (1953) 416-420.

22 R.A. Floyd and L.M. Soong, Spin trapping in biological systems. Oxidation of the spin trap 5,5- dimethyl-l-pyrroline-l-oxide by a hydroperoxide-hematin system, Biochem. Biophys. Res. Com- mun., 74 (1977) 79-84.

23 C. Mottley and R.P. Mason, Nitroxide radical adducts in biology: chemistry, applications and pitfalls. In: Biological Magnetic Resonance, L.J. Berliner and J. Reuben (Eds.), Plenum Publishing Corporation, Vol. 8, 1989, pp. 489-564.

24 K. Reszka, F.S. Cruz and R. Docampo, Photosensitization by the trypanocidal agent gentian violet. Type I versus Type II reactions, Chem.-Biol. Interact., 58 (1986) 161-172.

25 P. Moldeus, P. J. O'Brien, H. Thor, M. Berggren and S. Orrenius, Oxidation of glutathione by free radicals intermediates formed during peroxidase-catalyzed N-demethylation reactions, FEBS Lett., 162 (1983) 411-415.

26 K. Asada and S. Kanematsu, Reactivity of thiols with superoxide radicals, Agric. Biol. Chem., 40 (1976) 1891-1892.

27 M. Lal, 6°Co-radiolysis of reduced glutathione in aerated solutions at pH values between 1 - 7. Can J. Chem., 54 (1976) 1092-1097.

28 H.H. Young and J.H. Hill, The treatment of septicemia and local infections by the i.v. injection of mercurochrome 220-soluble and of gentian violet, J. Am. Med. Assoc., 82 (1924) 669-672.

29 J.P. Uetrecht, Idiosyncratic drug reactions: possible role of reactive metabolites generated by leukocytes, Pharm. Res., 6 (1989) 265-273.

30 B.H.J. Bielski, H.W. Richter and P.C. Chan, Some properties of the ascorbate free radical, Ann. N.Y. Acad. Sci., 258 (1975) 231-237.

31 B. Frei, B., L. England and B.N. Ames, Ascorbate is an outstanding antioxidant in human blood plasma, Proc. Natl Acad. Sci. U.S.A., 86 (1989) 6377-6381.

32 R. Docampo, S N.J, Moreno and F.S. Cruz, Enhancement of the cytotoxicity of crystal violet against Trypanosoma crazi in the blood by ascorbate, Mol. Biochem. Parasitol., 27 (1988) 241 - 247.