FreeRadicalBiology& Medicine, Vol. 15, pp. 159-167, 1993 Printed in the USA. All rights reserved.

0891-5849/93 $6.00 + .00 Copyright 1993 PergamonPress Ltd.

Original Contribution
IRON-REDUCING AND FREE-RADICAL-SCAVENGING PROPERTIES APOMORPHINE AND SOME RELATED BENZYLISOQUINOLINES OF

AMALIA UBEDA, CARMEN MONTESINOS, MIGUEL PAYh,, and MARIA JOSE ALCARAZ Departamento de Farmacologia, Facultad de Farmacia, Avda Vicent Andr6s Estell6s, s/n 46100 Burjassot, Valencia, Spain (Received 11 November 1992; Revised 3 February 1993; Accepted 26 February 1993) Abstract--The scavenging and iron-reducing properties of a series ofbenzylisoquinolinesof natural and synthetic origin have been studied. Bulbocapnine, boldine, glaucine, and stepholidine acted as scavengers of hydroxyl radical in the deoxyribose degradation by Fe3+-EDTA + H202. On the contrary, laudanosoline, apomorphine, protopapaverine, anonaine, and tetrahydroberberine increased deoxyribose degradation by a mechanism related to generation ofsuperoxide anion. Only apomorphine had a stimulating effect in the system using citrate instead of ethylenediaminetetraaceticacid (EDTA) as well as in the absence of chelator. Apomorphine also stimulated DNA damage by Cu 2+. The iron-ion reducing ability ofapomorphine and laudanosoline was confirmed using cytochrome c. Both compounds scavenged peroxyl radicals in an aqueous medium, while in Fea+-induced microsomal lipid peroxidation apomorphine acted as an inhibitor and laudanosoline stimulated the process. It is suggested that in microsomes the chain-breaking antioxidant properties of apomorphine overcome its possible influence on redox cycling of iron, or prooxidant properties. Keywords--Radical scavenger, Iron reducer, Apomorphine, Isoquinolines, Aporphines, Protoberberines, Microsomal lipid peroxidation, Free radicals

INTRODUCTION

It is well recognized that iron-mediated reactions are involved in the initiation and/or progression of lipid peroxidation and that reactive oxygen species play an important role in the etiology of a number of pathological statesY 2 Recent studies have indicated that antioxidant molecules protect against free-radical-mediated toxicity3-5 and thus they may be good candidates for moderating such pathologies. Many natural products are endowed with antioxidant properties, which are relevant not only from a phytochemical point of view concerning their role as secondary metabolites in plants, but also in relation to their nutritional incidence or as potential drugs of therapeutical or experimental application. The benzylisoquinoline alkaloids are comprised of a large group of secondary plant metabolites which display a variety of pharmacological actions. Other benzylisoquinolines of synthetic origin have also found therapeutical application. 6-s For example, apomorphine is a D2 dopaminergic agonist used as an
Address correspondence to: Prof. Maria Jose Alcaraz. 159

emetic, glaucine has a centrally acting antitussive activity, and boldine is the main active component of Boldo leaves, which are used for their stimulant action on the liver and diuretic properties. There is considerable variation in structure between the groups of isoquinoline alkaloids as the result of different stages in the common biogenetical pathway using tyrosine as precursor. Nevertheless, many of them possess phenolic or other reactive groups which suggest their possible participation in redox reactions. There have been few studies which have examined the effect of benzylisoquinoline derivatives on freeradical-mediated processes. Some biscoclaurine alkaloids (bisbenzylisoquinoline derivatives) are able to inhibit superoxide production by phagocytic cells and lipid peroxidation. 9-13 The inhibitory effect of boldine, a singlet oxygen quencher weaker than glaucine, 14 on brain homogenate autooxidation and peroxidation of red cell plasma membranes induced by 2,2'-azo-bis (2-amidinopropane), as well as lysozyme inactivation by peroxyl radicals, 15 has also been reported. The compounds included in this study (Fig. 1) be-

160

A. UBEDAd ul.

2/

11

'O

P
I
N\
\ I / 9

3/ 2\ I
14 13 / I 9

N
8

0%

'0

I 2 Apomorphine H H Boldiie OH 0CH3 Glaucine 0CH3 OCH3 AllOhlC -O-CH2-OBulkcapnine -0-CHz-o-

6 CH3 CH3 CH3 H CH3

9 H OH OCH3 H H

10 OH OCH3 0CH3 H 0CH3

11 OH H H H H

Protoberberine 2 3 Stcpholidiie 0CH3 OH Tetrahydroberherine -0-CH_2--O-

10

OH 0CH3

0CH3 0CH3

Fig. 1. Structure of the benzylisoquinolines be&ylisoquinoline).

tested. A = laudanosoline (simple benzylisoquinoline).

B = protopapaverine (simple

long to the simple benzylisoquinoline, aporphine, and protoberberine groups, which were selected in a previous screening based on their inhibitory activity on microsomal lipid peroxidation, where some of them demonstrated a potency comparable to known antioxidants.i6 In the present report we assessed both anti- and prooxidant effects of benzylisoquinoline derivatives, protopapaverine, laudanosoline, apomorphine, boldine, glaucine, anonaine, bulbocapnine, stepholidine, and tetrahydroberberine in several free-radical-mediated reactions in vitro in order to gain more insight into their mechanisms of action.
MATERIALS AND METHODS

Apomorphine, boldine, berberine, bulbocapnine, and glaucine were purchased from Sigma Chemical Co. (St. Louis, MO), protopapaverine from Roth GmBH (Karlsruhe, Germany), and laudanosoline from Aldrich Chemie (Steinheim, Germany). Alkaloids were used in salt form and dissolved in distilled water freshly before use. 2,2-Azo-his-(-2-amidinopropane) dichloride was obtained from Park Scientific Ltd. (Northampton). Cytochrome c type VI, catalase from bovine liver, DNA type I, superoxide dismutase (SOD) from bovine erythrocytes, and the rest of the chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).
Hydroxyl radical generation

Chemicals

Stepholidine and anonaine were isolated from Annona cherimolia stem bark. Tetrahydroberberine was obtained by Clemensen reduction of berberine.

Tubes containing 20 PM FeCl, , 1.4 mM H202, 2.8 mM deoxyribose, chelator, EDTA, or citrate (100 PM), and test compound, in 1 ml KH2P0,-KOH buffer, pH 7.4 ( 10 mM), were incubated in triplicate

Benzylisoquinolines as reducers

161

for 60 min at 37C. Solutions of iron salt, H202,alkaloid, and ascorbate were made up just before use and the iron salt was mixed with chelator before addition. Deoxyribose degradation by generated hydroxyl radical ("OH) was measured using the thiobarbituric acid method, t8 Test compounds which were dissolved in distilled water did not interfere with this detection method in control experiments.

Data analysis
Statistical analysis was performed using the Dunnews t test for multiple comparisons. Inhibitory concentration 50% (IC50) was calculated from the concentration/effect regression lines.
RESULTS

Hydroxyl radical generation DNA degradation

Reaction mixtures (1 ml) contained test compounds, 200 ~g calf thymus DNA, NaH2PO4-0.15 M NaC1, pH 7.4, 100 ~M FeCI3, or 200 #M CuSO4. Tubes were incubated in triplicate for 2 h at 37C and DNA degradation was estimated by the thiobarbituric acid method as above. Deoxyribose degradation can be influenced by benzylisoquinoline derivatives in a number of ways. Laudanosoline, protopapaverine, apomorphine, anonaine, and tetrahydroberberine dose dependently stimulated deoxyribose degradation by Fe3+-EDTA + H202 (Fig. 2). This reaction is inhibited by SOD, catalase, and mannitol, and stimulated by the iron reducer ascorbate (Table 1). The stimulatory effect of benzylisoquinoline derivatives is not additive to that of ascorbate and can be inhibited by the aforementioned enzymes and "OH scavenger. On the contrary, bulbocapnine, boldine, glaucine, and to a lesser extent stepholidine can behave as "OH scavengers in this system (Fig. 3). In the absence of H202 (Table 2), the majority of the compounds did not stimulate Fe3+-EDTA deoxyribose degradation. After performing the appropriate controls, we observed that laudanosoline and apomorphine were the only compounds able to stimulate deoxyribose degradation. For both benzylisoquinolines the reaction was inhibited by catalase and mannitol; nevertheless, SOD increased the effect of apomorphine and had the opposite influence in the case of laudanosoline. We have also tested benzylisoquinoline derivatives on "OH generation using as a chelating agent citrate instead of EDTA. In this system apomorphine stimulated deoxyribose degradation, but some compounds which enhanced "OH generation by Fe3+-EDTA + H202 significantly inhibited the reaction induced by Fe3+-citrate + H 2 0 2 , and the higher activity was found with laudanosoline (Fig. 4). The stimulatory effect of apomorphine was enhanced by SOD and reduced by catalase and mannitol. It is interesting to note that in the presence of ascorbate the resulting effect was lower than that of apornorphine or ascorbate separately (data not shown). When H202 was omitted from the incubation medium, Fe3+-citrate did not induce "OH generation. Nevertheless, a significant generation was observed after addition of apomorphine. Again, the reaction was inhibited by catalase and mannitol and stimulated by SOD (data not shown).

Cytochrome c reduction
The reduction of cytochrome c (0.1 mM) in 10 mM KH2PO4-KOH buffer, pH 7.4, with 30 ~M EDTA was followed for 2 min at 25C and 550 nm in the presence and in the absence of test compounds.

Microsomal lipid peroxidation induced by Fe 3+

Rat liver microsomal fractions were prepared as
d e s c r i b e d 19 from livers of male Wistar rats weighing

200-250 g. Reaction mixtures contained 1.7 mg microsomal protein/ml, 200 t~M FeCl3 6H20, NaCI (0.9%) adjusted to pH 7.4 before use, and test compounds. The samples were incubated in triplicate at 37C for 30 min, and the extent of lipid peroxidation was determined by the thiobarbituric acid method. 2

Peroxyl radical scavenging

Reaction mixtures (1 ml) contained the following: 50 mM KH2PO4-KOH pH 7.4, 0.68 mM lysozyme, 10 mM 2,2'-azo-bis-(2-amidinopropane)dihydrochloride, and test compounds at different concentrations. The thermal decomposition of the azo initiator was carried out at 45C for 90 min. Aliquots of 50 #1 were added to 950 #1 of a suspension ofMicrococcus lysodeikticus (0.3 mg/ml) in Dulbecco's buffer. The change ofabsorbance was followed at 450 nm during the first minute. 2~ Control experiments showed that test compounds did not interfere with lysozyme activity.

162
2.4

A. UBEDA el al.

1.8

Ass 2

1.2

0.11

A1

A2

A3

AP1 AP2 AP3

L1

L2

L3

P1

P2

P3

T1

12

13

Fig. 2. Effect ofstimulatory benzylisoquinolines on deoxyribose degradation by Fe3+-EDTA + H202. Results show mean _+ SE of absorbance units from 6-24 determinations. C - control, A l* - anonaine 1 tzM, A2** = anonaine 10 uM, A3** = anonaine 100/~M, API** = apomorphine 1 ~M, AP2** = apomorphine 10/aM, AP3** - apomorphine 100 ~M, L1 ** = laudanosoline 1 ~M, L2** = laudanosoline 10 uM, L3** - laudanosoline 100 ~M, PI* - protopapaverine 1 tzM. P2** - protopapaverine 10 uM, P3** - protopapaverine 100 ttM, TI** - tetrahydroberberine 1 ~tM, T2** - tetrahydroberberine 10 uM, T3** = tetrahydroberberine 100 uM. *p < .05, **p < .01.

In the absence of chelator, the deoxyribose assay is a way to assess the ability of a compound to interfere with site-specific generation of "OH radicals. 22 As shown in Table 3, none of the alkaloids tested inhibited this reaction as did desferoxamine, but apomorphine increased deoxyribose degradation by Fe 3. Stimulatory benzylisoquinolines were assayed for their ability to modify DNA degradation induced by Cu 2. In other experiments we used F e 3+ instead of Cu 2, with the result of very reduced absorbances (data not shown). Only apomorphine exerted significant effects with an increase in DNA degradation by

Cu 2+ of about 80%, which was unaffected by catalase and slightly inhibited by mannitol, while SOD increased the extent of DNA degradation by Cu 2+ + apomorphine (Table 4).
Cytochrome c reduction

Since the experiments on the generation of "OH radical suggested the interaction of some benzylisoquinolines with iron ions, we confirmed the direct effects of this series of benzylisoquinolines on cytochrome c. From the data (Fig. 5) we conclude that

Table 1. Effect of Benzylisoquinolines on Deoxyribose Degradation by Fe3+-EDTA + H202 (Influence of SOD, Catalase, Mannitol, and Ascorbate) Fe3+.EDTA + H202 Control Laudanosoline Protopapaverine Apomorphine Bulbocapnine Boldine Glaucine Anonaine Tetrahydroberberine Stepholidine 0.145 1.067 0.335 2.240 0.085 0.089 0.068 0.318 +_ 0.002 _+ 0.016 _+ 0.008 + 0.034 _+ 0.003 _+ 0.002 _+ 0.002 +_ 0.010 (77) (23)** (23)** (23)** (13)** (17)** (16)** (24)**

+ S O D (100 U) 0.079 0.279 0.109 1.695 0.056 0.062 0.055 0.109 + 0.002 _+ 0.006 _+ 0.002 _+ 0.076 _+ 0.003 +_ 0.001 +_ 0.001 + 0.003 ( 12)** (6) ++ (6) ++ (6) ++ (3) ++ (3) ++ (3) + (6) ++

+Catalase (100 U) 0.015 0.024 0.025 0.031 0.007 0.008 0.011 0.026 _+ 0.001 _+ 0.002 _+ 0.001 _+ 0.004 _+ 0.001 +_ 0.002 _+ 0.001 _+ 0.001 ( 15)** (6) (6) ++ (6) ++ (3) ++ (3) + (3) (6) ++

+Mannitol (50 mM) 0.058 0.130 0.075 0.231 0.035 0.026 0.041 0.059 _+ 0.002 _+ 0.004 _+ 0.003 + 0.010 _+ 0.002 _+ 0.001 _+ 0.001 _+ 0.003 ( 12)** (6) ++ (6) + (6) + (3) ++ (3) ++ (3) ++ (6) ++

+Ascorbate (50 ~M) 0.831 1.240 0.598 2.317 0.532 0.752 0.560 0.667 _+ 0.022 _+ 0.039 _+ 0.008 _+ 0.071 +_ 0.015 _+ 0.006 + 0.006 _+ 0.034 ( 15)** (6) ++ (6) ++0 (6) 0 (6) ++0 (6) ++ (5) ++0 (9) . . . .

0.729 _+ 0.012 (29)** 0.114 + 0.004 (17)**

0.091 + 0.004 (6) ++ 0.081 + 0.002 (3) ++

0.012 _+ 0.001 (6) + 0.022 _+ 0.001 (3) ++

0.085 _+ 0.004 (6) ++ 0.038 + 0.001 (3) ++

1.124 _+ 0.043 (6) ++0 0.621 _+ 0.010 (6) ++0

Results show mean _+ SE of absorbance units from (n) determinations. Benzylisoquinolines were tested at 100 ~M. * p < .05, ** p < .01, with respect to control Fe3+-EDTA + H202. + p < .05, ++ p < .01 with respect to benzylisoquinoline + Fe3+-EDTA + H202. oo p < .01 with respect to control Fe3+-EDTA + H202 + ascorbate.

Benzylisoquinolines as reducers
0.1Q

163

0.12

AS3 2

O.Oe

J_

I
0.04 ! 0 C 8U1 BU2 BU3 B01 B02 B03 (31 G2 03 $1 $2 $3

Fig. 3. Effect of inhibitory benzylisoquinolines on deoxyribose degradation by Fe3+-EDTA + H202. Results show mean _+ SE o f absorbance units from 6-24 determinations. C = control, Bu 1"* = bulbocapnine I , M , Bu2** = bulbocapnine 10/~M, Bu3** = bulbocapnine 100 , M , Bol** = boldine 1 tzM, Bo2** = boldine 10 ~tM, Bo3** = boldine 100 , M , GI** = glaucine 1 #M, G2** = glaucine 10 , M , G3** = glaucine 100/zM, SI** = stepholidine 1 uM, $2"* = stepholidine 10 #M, $3"* = stepholidine 100 /~M. **p < .01.

apomorphine and laudanosoline reduce Fe 3+ in this system, to a lesser extent than ascorbate. The rest of the benzylisoquinolines did not reduce cytochrome c (data not shown).

DISCUSSION

Deoxyribose degradation by Fe3+-EDTA + H202 is dependent on "OH generation in a Fenton reaction: Fe3+-EDTA
+

Microsomal lipid peroxidation induced by Fe 3

The results in this system are shown in Table 5. Apomorphine was the most effective inhibitor, followed by bulbocapnine, tetrahydroberberine, stepholidine, boldine, and glaucine. On the contrary, anonaine slightly stimulated peroxidation and laudanosoline increased it over 50%. Only protopapaverine was without effect.

H202
+ H202 ~

Intermediate Fe complex-Fe2-EDTA
+

Fe2+-EDTA

H202 --~ (see Ref. 18).

Fe3+-EDTA + .OH + OH-

Peroxyl radical scavenging

Laudanosoline and apomorphine were potent inhibitors of lysozyme inactivation by peroxyl radicals (Table 6) with ICso in the #M range.

Some of the aporphines tested--bulbocapnine, boldine, glaucine, and the protoberberine stepholidine--acted as scavengers of "OH in this system. The rest of the compounds--laudanosoline, apomorphine, protopapaverine, anonaine, and tetrahydroberberine--increased deoxyribose degradation, which was significantly inhibited by catalase and to a lesser extent by mannitol. These results are consistent

Table 2. Efect of Laudanosoline and Apomorphine on Deoxyribose Degradation by Fe3-EDTA (Influence o f SOD, Catalase, Mannitol, and Ascorbate) Fe3+-EDTA Control Laudanosoline Apomorphine 0.005 _+ 0.001 (18) 0.141 _+ 0.006 (9)** 0.157 _+ 0.007 (9)** +SOD (100 U) 0.010 _+ 0.002 (8) 0.040 _ 0.003 (6) + 0.458 + 0.028 (5) + +Catalase (100 U) 0.010 + 0.002 (7) 0.012 _+ 0.000 (6) ++ 0.016 + 0.002 (6) + +Mannitol (50 mM) 0.007 _+ 0.001 (5) 0.013 + 0.001 (6) ++ 0.021 _+ 0.001 (6) ++ +Ascorbate (50 , M ) 0.206 _+ 0.009 (7)** 0.248 _+ 0.003 (6) ++ 0.285 _+ 0.012 (6) ++

Results show mean + SE of absorbance units from (n) determinations. Benzylisoquinolines were tested at 100/~M. ** P < .01, with respect to control Fe3-EDTA. ++ P < .01 with respect to benzylisoquinoline + Fe3+-EDTA. P < .01 with respect to control Fea-EDTA + ascorbate.

164
0.4

A. UBEDA et al.

0.3

As32 0.2

0.1

AP

BU

I10

Itll
T

Fig. 4. Effect ofbenzylisoquinolines on deoxyribose degradation by Fe3+-citrate + H202. Results show m e a n _+ SE ofabsorbance units from 6 - 9 determinations. Benzylisoquinolines were tested at 100 uM. C = control, L** = laudanosoline, P** = protopapaverine, AP** = apomorphine, Bu** = bulbocapnine, Bo** = boldine, G** = glaucine, A = anonaine, T** = tetrahydroberberine, S** = stepholidine. **p < .01.

with a stimulation of "OH production by these alkaloids. In this regard, a clear structure/activity relationship has not been observed. In control incubations, addition of ascorbate greatly accelerated "OH generation due to Fe 3+ reduction, 18and we observed that deoxyribose degradation in the FeX+-EDTA + H202 + ascorbate system was also inhibited by bulbocapnine, boldine, glaucine, and stepholidine. It is interesting to note that protopapaverine and anonaine, which moderately increased "OH formation by Fea+-EDTA + H202, inhibited deoxyribose degradation when ascorbate was present, while stimulatory benzylisoquinolines, apomorphine, laudanosoline, and tetrahydroberberine induced an

increase in absorbance in the presence of ascorbate lower than expected, suggesting a competition between these benzylisoquinolines and ascorbate in redox cycling the iron. SOD inhibited "OH generation by Fe3+-EDTA + H202 in control incubations and when incubations were performed in the presence ofstimulatory benzylisoquinolines. A possible explanation would be related to dismutation of 02"- generated during benzylisoquinoline oxidation, thus preventing the superoxide-driven Fenton reaction with generation of " O H : 22 Benzylisoquinoline + O2 -+ Oxidized benzylisoquinoline + O2"---SOD ~ H202

Table 3. Effect o f Benzylisoquinolines on Deoxyribose Degradation by Fe > Control Laudanosoline Protopapaverine Apomorphine Bulbocapnine Boldine Glaucine Anonaine Tetrahydroberberine Stepholidine 0.089 0.077 0.087 0.190 0.091 0.094 0.102 0.100 0.080 0.085 _+ 0.005 + 0.006 + 0.004 _+ 0.003 _+ 0.005 + 0.011 _+ 0.010 + 0.008 _+ 0.005 _+ 0.004 (16) (9) (6) (6)** (6) (6) (6) (6) (9) (9)

02"- + Fe3+-EDTA + Fe2+-EDTA + 02

H20 2

q- Fe2+-EDTA --+ Fe3+-EDTA + .OH + OH-.

Results show m e a n +_ SE of absorbance units from (n) determinations. C o m p o u n d s were tested at 100 , M . In this system the value for the reference desferoxamine was 0.058 _+0.001 (6)**. ** p < .01.

As a result, dismutation of 02"- by SOD significantly attenuated the reduction of iron ions by benzylisoquinolines. This mechanism could operate for all the stimulatory compounds included in our study. Nevertheless, in the presence of apomorphine, the most effective stimulant, we observed a modest reduction in absorbance by SOD, suggesting that 02"generation plays a minor role in the mechanism of "OH overproduction by this aporphine. On the other hand, in the absence of exogenous H202, the Fenton reaction did not take place in con-

Benzylisoquinolines as reducers Table 4. Effect of Apomorphine, Laudanosoline, and Tetrahydroberberine on DNA Degradation by Cu2 (Influence of SOD, Catalase, and Mannitol) Cu2(200 ,M) Control Laudanosoline Apomorphine Tetrahydroberberine 0.047 +_0.006 (7) 0.071 _+0.010 (6)** 0.085 +_0.005 (6)** 0.036 + 0.003 (6) +SOD (100 U) 0.056 _+0.001 (6) 0.068 _+0.012 (6) 0.231 +_0.007 (6)++ 0.045 _+0.001 (6) +Catalase (100 U) 0.050 _+0.003 (6) 0.080 _+0.006 (6) 0.100 _+0.011 (6) 0.035 _+0.002 (6)

165

+Mannitol (50 mM) 0.031 + 0.002 (6) 0.070 _+0.003 (6) 0.069 _+0.010 (6) 0.030 +_0.004 (6)

Results show mean _+SE of absorbance units from (n) determinations. Compounds were tested at 100 #M. ** p < .0 l, with respect to control Cu2+. ++p < .01 with respect to apomorphine + Cu2+.

trol incubations, but when a p o m o r p h i n e or laudanosoline were added the generation of "OH from H 2 0 2 was apparent, resulting in deoxyribose degradation. In the case o f a p o m o r p h i n e addition o f SOD to this incomplete system (without H202) significantly enhanced deoxyribose degradation, which might be due either to an increase in a p o m o r p h i n e oxidation in the presence o f SOD or to the contribution ofO2"- dismutation to H 2 0 2 production. Recently the importance of the chelator used in the generation of "OH by Fenton-type reactions has been pointed out, since agents such as adenosine 5'-triphosphate (ATP) or citrate do not allow the redox cycling of iron and are considered to be m o r e relevant than E D T A in physiological situations. Thus Fe3+-citrate + H 2 0 2 would provide a convenient assay for "OH scavengers. 23 In experiments using citrate as chelator, we obtained a lower rate of "OH generation, which was increased only by apomorphine. Inhibition by cat-

alase and mannitol indicated the participation o f H 2 0 2 and "OH in this reaction in control incubations or in the presence of apomorphine, while addition o f SOD further stimulated deoxyribose degradation by this aporphine, although this effect was lower than that o f the Fe3+-EDTA complex. In the incomplete system (without H202) a p o m o r p h i n e was able to induce the Fenton reaction, which was stimulated by addition o f SOD. Laudanosoline and tetrahydroberberine, which enhanced deoxyribose degradation by Fe3-EDTA + H 2 0 2 , might show "OH scavenging activity in situations where their action on iron redox is affected, as in the system using citrate instead of EDTA. The rest o f the benzylisoquinolines, except anonaine, behaved as "OH scavengers in the system Fe3+-citrate. Some of the benzylisoquinolines tested possess chemical groups such as orthodihydroxyl groups, which might bind iron ions. Our results show that

0.6

0.4

O.3
t

0.2

oJ
0

r-q
i
. . . .

Fig. 5. Effectofapomorphine and laudanosoline on cytochrome c. Results show mean _+SE ofabsorbance increaserate from six determinations. API** = apomorphine 1 ,M, AP2** = apomorphine 10 gM, AP3** = apomorphine 100/zM, LI = iaudanosoline 1 ~aM, L2** = laudanosoline 10 ~M, L3** = iaudanosoline 100 ,M, ASI** = ascorbate 1 t~M, AS2** = ascorbate 10 uM, AS3** = ascorbate 100 ,M. **p < .01.

166

A. UBEDA et al. Table 6. Effect of Laudanosoline and Apomorphine on Peroxyl Radical-Induced Inactivation of Lysozyme Compound Laudanosoline Apomorphine % Inhibition at 100 ~M 100.0 _+ 0.0"* 91.6 _+ 3.9** IC5o (uM) 5 12

these benzylisoquinolines do not inhibit site-specific generation of "OH radicals, suggesting that the chelation of iron ions is not the mechanism of their inhibitory action on "OH radical generation. Apomorphine was the only compound which enhanced Fe3+-in duced "OH radical generation, indicating its ability to stimulate iron ion-dependent site-specific Fenton chemistry. It is known that DNA damage due to oxidative stress in vivo is related to "OH production after release of iron or copper ionsfl 4 Apomorphine can reduce Cu 2+, thus increasing DNA damage by this ion, and the stimulatory effect of SOD on DNA damage by apomorphine + Cu 2+ is consistent with the results obtained for this benzylisoquinoline in iron-mediated reactions. It is recognized that apomorphine is oxidized by air in neutral solutions to the colored quinone derivative, 25 a process which could be catalyzed by Fe 3+, Fe3+-EDTA, Fe3+-citrate, or Cu 2+, resulting in oxygen radical formation with possible deleterious effects in biological molecules. The reducing ability ofapomorphine and laudanosoline was further confirmed in the cytochrome c experiments. The latter compound showed an effect closer to that ofascorbic acid at the highest concentration tested (100 uM). The stimulation of Fe3+-in duced lipid peroxidation by laudanosoline was expected due to its iron-reducing ability. In this regard laudanosoline would behave like ascorbic acid. Nevertheless, apomorphine strongly inhibited lipid peroxidation in microsomes. We have demonstrated the "OH scavenging properties of a number of benzylisoquinolines, although as reported previously for known "OH scavengers 26 we did not observe any relationship between scavenging

Results show mean _+ SE of percentages of inhibition from six tests. For IC5o determination a range of five concentrations was used. ** p < .01.

activity and inhibition of Fe3+-stimulated lipid peroxidation. Our present data indicate that both apomorphine and laudanosoline are scavengers of peroxyl radicals in an aqueous medium, with a potency similar to that showed for inhibition of FeZ+/cysteine-induced lipid peroxidation. 16We previously observed (unpublished work) that apomorphine but not laudanosoline is consumed during the lipid peroxidation process, suggesting an interaction with peroxyl radicals in microsomes. It is interesting to note that apomorphine is more lipophilic than laudanosoline, a feature that could facilitate its access to propagating radicals in membranes during the peroxidative process. Our present results thus suggest that in microsomes the chainbreaking antioxidant properties of apomorphine overcome its influence on redox cycling of iron, or prooxidant properties.
Acknowledgements - - This work was supported by grant PM900126 from CICYT, Spanish Ministerio de Educaci6n y Ciencia. M.CM. thanks Consellerla de Cultura, Educaci6n y Ciencia, Generalitat Valenciana, for a fellowship.

REFERENCES 1. Yagi, K. Lipid peroxides and human diseases. ('hem. Phys. Lipids 45:337-351; 1987. 2. Pollack, R. L.; Morse, D. R. Free radicals and antioxidants: Relation to chronic diseases and ageing. Int. J. Psychosom. 35:43-48; 1988. 3. Liu, G.-T.; Zhang, T.-M.; Wang, B.-E.; Wang, Y.-W. Protective action of seven natural phenolic compounds against peroxidative damage to biomembranes. Biochem. Pharmacol. 43:147-152; 1992. 4. Pritsos, C. A.; Sokoloff, M.; Gustafson, D. U PZ-51 (Ebselen) in vivo protection against Adriamycin-induced mouse cardiac and hepatic lipid peroxidation and toxicity. Biochem. Pharmacol 44:839-841 ; 1992. 5. Payfi, M.; Ferrfindiz, M. L.; Sanz, M. J.; Alcaraz, M. J. Effects of phenolic compounds on bromobenzene-mediated hepatotoxicity. Xenobiotica; 1993. In press. 6. Sime6n, S.; Rios, J. L.; Villar, A. Pharmacological activities of benzophenanthridine and phenanthrene alkaloids. Pharmazie 44:593-597; 1989. 7. Rios, J. L.; Sime6n, S.: Villar, A. Pharmacological activity of

Table 5. Effect of Benzylisoquinolines on Microsomal Lipid Peroxidation Induced by Fe 3+ Compound Laudanosoline Protopapaverine Apomorphine Bulbocapnine Boldine Glaucine Anonaine Tetrahydroberberine Stepholidine % of Control 157.6 92.4 0.0 43.6 69.8 86.4 119.0 53.2 65.4 _+ 3.2 (7)** 1.3 (9) 0.1 (6)** _+ 1.6 (6)** _+ 2.1 (6)** _+ 2.3 (13)** _+ 4.7 (5)** _+ 1.0 (6)** _+ 4.2 (10)**

Results show mean _+ SE of percentages of control group from (n) determinations. Compounds were tested at 100 uM. **p < .01.

Benzylisoquinolinesas reducers aporphinoid alkaloids. A review. Fitoterapia 60:387-412; 1989. Sime6n, S.; Rios, J. L.; Villar, A. Pharmacological activities of protoberberine alkaloids. PI. Med. Phytothdr. 23:202-250; 1989. Shiraishi, N.; Arima, T.; Aono, K.; Inoue, B.; Morimoto, Y.; Utsumi, K. Inhibition by biscoclaurine alkaloid of lipid peroxidation in biological membranes. Physiol. Chem. Physics 12:299-305; 1980. Haisong, J.; Xiaojie, L.; Baolu, Z.; Zhewu, H.; Wenjuan, X. Scavenging effect of berbamine on active oxygen radicals in phorbol ester-stimulated human polymorphonuclear leukocytes. Biochem. Pharmacol. 39:1673-1678; 1990. Matsuno, T.; Orita, K,; Sato, E.; Nobori, K.; Inoue, B.; Utsumi, K. Inhibition of metabolic response of polymorphonuclear leukocyte by biscoclaurine alkaloids. Biochem. Pharmacol. 36:1613-1616; 1987. Matsuno, T.; Orita, K.; Edashige, K.; Kobuchi, H.; Sato, E. F.; lnouye, B.; Inoue, M.; Utsumi, K. Inhibition of active oxygen generation in guinea-pig neutrophils by biscoclaurine alkaloids. Biochem. Pharmacol. 39:1255-1259; 1990. Castranova, V.; Kang, J. H.; Moore, M. D.; Pailes, W. H.; Frazer, D. G.; Schwegler-Berry, D. Inhibition of stimulant-induced activation of phagocytic cells with tetrandrine. J. Leukoc. Biol. 50:412-422; 1991. Larson, R. A.; Marley, K. A. Quenching ofsinglet oxygen by alkaloids and related nitrogen heterocycles. Phytochemistry 23:2351-2354; 1984. Speiski, H.; Cassels, B. K.; Lissi, E. A.; Videla, L. A. Antioxidant properties of the alkaloid boldine in systems undergoing lipid peroxidation and enzyme inactivation. Biochem. Pharmacol. 41:1571-1581; 1991.

167

8. 9.

10.

11.

12.

13.

14. 15.

16. Martinez, L. A.; Rios, J. L.; Payfi, M.; Alcaraz, M. J. Inhibition of nonenzymic lipid peroxidation by benzylisoquinolinealkaloids. Free Radic. Biol. Med. 12:287-292; 1992. 17. Sime6n, S.; Rios, J. L.; Villar, A. Alkaloids from Annona cherimolia (Mill.) stem bark. PI. Med. Phytothdr. 23:159-161 ; 1989. 18. Laughton, M. J.; Halliwell, B.; Evans, P. J.; Hoult, J. R. S. Antioxidant and pro-oxidant actions of the plant phenolics quercetin, gossypol and myricetin. Effects on lipid peroxidation, hydroxyl radical generation and bleomycin-dependent damage to DNA. Biochem. Pharmacol. 38:2859-2865; 1989. 19. Slater, T. F.; Sawyer, B. C. The stimulatory effects of carbon tetrachloride and other halogenoalkanes on peroxidative reaction in rat liver fractions in vitro. Biochem. J. 123:805-814; 1971. 20. Braughler, J. M.; Duncan, L. A.; Chase, R. L. The involvement of iron in lipid peroxidation. Importance of ferric to ferrous ratios in initiation. J. Biol. Chem. 261:10282-10289; 1986. 21. Lissi, E. A.; Clavero, N. Inactivation of lysozyme by alkylperoxyl radicals. Free Radic. Res. Comm. 10:177-184; 1990. 22. Halliwell, B. How to characterize a biological antioxidant. Free Radic. Res. Comm. 9:1-32; 1990. 23. Puppo, A. Effect of flavonoids on hydroxyl radical formation by Fenton-type reactions; influence of the iron chelator. Phytochemistry 31:85-88; 1992. 24. Halliwell, B.; Aruoma, O. I. DNA damage by oxygen-derived species. Its mechanism and measurement in mammalian systems. FEBS Lett. 281:9-19; 1991. 25. Shamma, M. The isoquinoline alkaloids. Chemistry and pharmacology. New York and London: Academic Press; 1972. 26. Omar, R. F.; Hasinoff, B. B.; Mejilla, F.; Rahimtula, A. D. Mechanism of ochratoxin A stimulated lipid peroxidation. Biochem. Pharmacol. 40:1183-1191; 1990.