Boldine and its antioxidant or health-promoting properties
Peter O’Briena, Catalina Carrasco-Pozob, Hernan Speiskyb,c,∗a Graduate Department of Pharmaceutical Sciences, Faculty of Pharmacy, University of Toronto, Toronto, Ont., Canada
b Micronutrients Unit, Nutrition and Food Technology Institute, University of Chile, El Lıbano 5524, Macul,P.O. Box 138-11, Santiago, Chile
c Pharmacological and Toxicological Chemistry Department, Faculty of Chemical and Pharmaceutical Sciences,University of Chile, Santiago, Chile
Received 17 August 2005; received in revised form 8 September 2005; accepted 9 September 2005Available online 10 October 2005
Abstract
The increasing recognition of the participation of free radical-mediated oxidative events in the initiation and/or progression ofcardiovascular, tumoural, inflammatory and neurodegenerative disorders, has given rise to the search for new antioxidant molecules.An important source of such molecules has been plants for which there is an ethno-cultural base for health promotion. An importantexample of this is boldo (Peumus boldus Mol.), a chilean tree whose leaves have been traditionally employed in folk medicine and isnow widely recognized as a herbal remedy by a number of pharmacopoeias. Boldo leaves are rich in several aporphine-like alkaloids,of which boldine is the most abundant one. Research conducted during the early 1990s led to the discovery that boldine is one of themost potent natural antioxidants. Prompted by the latter, a large and increasing number of studies emerged, which have focused on
(S)-2,9-Dihydroxy-1,10-dimethoxy-aporphine (Bol-dine) and other related aporphine alkaloids have beenshown to behave as potent antioxidants in a num-ber of experimental models. Pharmacological activities,such as cyto-protective, anti-tumour promoting, anti-inflammatory, antipyretic and antiplatelet, have beenassociated with the ability of boldine to scavenge highlyreactive free radicals. The occurrence and chemistryof boldine and closely related congeners will first bebriefly recalled. The present review will address anddiscuss results from studies conducted mostly over thelast decade on the antioxidant activity of boldine, andwhen relevant, of some structurally related compounds,expanding into the main pharmacological properties,which arise from such activity. In addition, it will attemptto cover data supporting pharmacological properties ofthese substances, which are not necessarily related to itsantioxidant activity.
1.1. Occurrence and chemistry of boldine,derivatives and congeners
Mapuche verbs “weltum” (to sprout again) or “volitum”(to put out new roots), which may refer to this fea-ture. The indications for the use of boldo are extremelybroad in range and just as unsubstantiated. Accordingto pharmacopoeias and treatises dealing with medicinalplants, boldo extracts have been used for the treatmentof headache, earache, rheumatism, “nervous weakness”,dropsy, dyspepsia, menstrual pain, urinary tract inflam-mation and has also been claimed to be a sedative andmild hypnotic[1,2]. Boldo leaves contain between 0.4and 0.5% of at least 17 different alkaloids belongingto the large benzylisoquinoline-derived family[3]. Bol-dine is the major alkaloid as it accounts for 12–19% ofthe total alkaloid content[4]. Recently, Quezada et al.[5] estimated boldine content in boldo leaves as 0.14%,a value, which is similar to the 0.12% value previouslyreported by us[6]. Boldo leaves contain tannins, essen-tial oils (mainly ascaridole and cineole) and flavonoids,of which catechin was recently reported to be the mostabundant one[5,7]. Since the contents of catechin, gal-lic acid and tannic acid in boldo are much higher thanthat of boldine, the former compounds appear to con-tribute by a greater extent to the total antioxidant capacity
y%-here,
Boldine (Table 1), is the major leaf and bark alka-loid of the Chilean boldo tree. The boldo tree (Peu-mus boldus Molina, Monimiaceae) grows abundantlyin the more humid ecosystems of the Mediterranean
of boldo infusions[5,7]. Boldo bark contains unusuallhigh concentrations of boldine, with yields of up to 6based on dry weight[8]. Besides boldine, boldo contains, although in much smaller amounts, several otstructurally related alkaloids, amongst which glaucin
climatic region of central Chile and extends into thet,
s
an O-dimethylated form of boldine (Table 1) and theu aref -l
H3
3
3
H3
H3H3
3
H3)2
H3)2
nsaturated boldine analogue 6a,7-didehydroboldineound [2] (Fig. 1). Amongst the non-aporphinoid alkaoids of boldo, the benzyltetrahydroisoquinoline (R)-
northern half of the much rainier Chilean lake districbetween 33◦ and 39◦ South latitude. The name “boldo”or “boldu” is presumably derived from the indigenou
Table 1Structure of aporphines
Name 1 2 3 6
Boldine OCH3 OH H CGlaucine OCH3 OCH3 H CHIsoboldine OH OCH3 H CHBulbocapnine –O–CH2–O– H CAnonaine –O–CH2–O– H HApomorphine H H H C3-Bromo-boldine OCH3 OH Br C3-Iodo-boldine OCH3 OH I CHN-methylglaucinium OCH3 OCH3 H (CN-methylboldinium OCH3 OH H (C
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 3
Table 2Structure of benzylisoquinolines
Name 2 4 5 6 7
Reticuline CH3 OCH3 OH OCH3 OHCoclaurine H OH H OCH3 OHLaudanosine CH3 OCH3 OCH3 OCH3 OCH3
Laudanosoline CH3 OH OH OH OH
Fig. 1. 6a,7-Didehydroboldine.
reticuline (Table 2) is a common biosynthetic precursorof 1,2,9,10- and 1,2,10,11-tetraoxygenated aporphines.Boldo bark has also a considerable amount of (R)- and(S)-coclaurines (Table 2), which are common precur-sors of benzylisoquinoline-derived alkaloids. Althoughpresent as a minor constituent, boldine also occurs in anumber of other species of the Monimiaceae, Magno-liaceae and Lauraceae, along with many other aporphi-noids.
The analytical detection and quantification of boldine,early relied on the use of colorimetric, paper elec-trophoresis, voltammetric, thin layer and gas chromato-graphic methods. Based on the two absorption maxima(282 and 303 nm) of boldine, an HPLC method coupledto a UV detector was subsequently developed by Pietta etal. [9] to be applied to pharmaceutical preparations andby our laboratory[10] to assay boldine in plasma and inother biological fluids, making it possible to initiate phar-macokinetic studies. Using the latter method, we showedthat upon its oral administration to rats (50–75 mg/kg)boldine was rapidly absorbed and preferentially concen-trated in the liver[11].
1.2. Antioxidant activity of boldine
Research on natural antioxidants, prompted by thegrowing interest in free radical-induced biological dam-age, intensified in the 1980s and 1990s partly as a resultof the perceived need to screen endangered floras for
substances of potential therapeutic utility. Of particularinterest as possible sources of antioxidants were medic-inal plants traditionally used to treat conditions, whichare related to oxidative stress, such as rheumatism andinflammatory liver diseases[12,13]. Thus, the possibil-ities of preventing or retarding the deleterious effectsassociated with excessive production of reactive oxy-gen species (ROS), by the use of previously unexploredgroups of natural products seemed attractive as a sub-ject of research. This endeavour was further stimulatedby the unsubstantiated belief that their natural characterwould imply innocuousness and by increasing evidenceof the expression of some forms of toxicity both, afteracute and prolonged exposure to high doses of syntheticantioxidants like butylated hydroxytoluene (BHT) andbutylated hydroxyanisole (BHA)[14,15]. Conversely,the ready availability of flavonoids appeared then as apotential alternative to the synthetic compounds. How-ever, it was soon shown that some of these natural sub-stances too might pose risks in terms of mutagenicity(vide infra) and that limitations to their in vivo use areimposed by their low oral bioavailability, unfavourablepharmacokinetics and unknown toxicity concerns[16].
Our own interest in boldine arose from the observa-tion of structural analogies between this natural productand recognized antioxidant substances, namely struc-tures featuring phenolic hydroxyl groups. Such interestwas further prompted by the fact that nowadays boldine-containing herbal teas are widely consumed in SouthAmerica and boldo leaves are being continually exported
euti-Mostntsicalsroupbuten-tectar-
to some European countries for further pharmaccal processing to boldine-containing concentrates.well-known natural and synthetic phenolic antioxidaare sterically hindered phenols, so that the free radgenerated by hydrogen abstraction from the OH g(Fig. 2, reaction B) are not only thermodynamically,also kinetically stable or “persistent”, i.e., their propsity to react with the biological substrates they pro(reaction C) is substantially lower than that of the “pent” free radical (reaction A).
4 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
However, it was our early observations of the verypotent antioxidant activity of boldine in biological[17]and abiotic[18] systems that encouraged us to system-atically explore its actual potential as an antioxidant andcyto-protective agent. 2,2′-Azobis[2-amidinopropane](AAPH) is a water-soluble azo-compound whose ther-mal decomposition generates alkylperoxyl radicals. Weshowed initially that low concentrations of boldine(IC50 = from 5× 10−6 to 15× 10−6 M) effectively pro-tected red blood cell plasma membranes against theAAPH-induced increase in antioxidant-sensitive O2uptake[17]. The latter parameter is considered as anindex of the rate at which O2 reacts with polyunsaturatedfatty acid (PUFA)-derived free radicals, propagating theoxidation of membrane lipids. Noteworthy, in such a sys-tem the antioxidant (anti-propagation or chain-breaking)efficiency of boldine (Ki = 13�M) was found to be sim-ilar to that reported for the widely used bioflavonoid(+)-cyanidanol-3 and about 30-fold greater than thatof silybin, a mixture of lignano-flavonoids availablefor therapeutic use as an antioxidant[19]. Boldine wasalso demonstrated to effectively inhibit the spontaneousautoxidation of brain membrane lipids (assessed by pro-duction of thiobarbituric reactive substances or TBARS,O2 uptake and chemiluminescence) with an apparentKi of about 19–30�M [17]. Autoxidation of lipids isa process that mainly reflects the overall rate of tran-sition metal-dependent decomposition of lipid perox-ides, both, initially occurring and those formed duringthe incubation conditions. Shortly before, Rıos et al.
linex-o-entals onlka-
y,loidsed bypre-sha-hed
[23] that boldine acts as a very efficient HO• scavenger.Such an ability was subsequently confirmed by Janget al. [24] and by Youn et al.[25]. The latter investi-gators observed that boldine also inhibited nitric oxideproduction by the mitochondria. Yet, in contrast withthe suggestion by Rıos et al.[20], boldine was foundto either not react with superoxide radicals[23] or toreact very poorly with these species[24]. The lack ofsuperoxide-scavenging properties of boldine may be infact regarded as a possibly advantageous feature sincesuch radicals are physiologically generated by cells andmay constitute important intra- and intercellular medi-ators of essential biological processes. Kringstein andCederbaum[26] reported that boldine prevents the non-enzymatic peroxidation of microsomal lipids initiated byFe2+ or the enzyme-catalyzed peroxidation initiated byFe3+–ATP with NADPH or NADH as cofactors or ini-tiated by CCl4 plus NADPH as cofactor. Interestingly,boldine was shown to prevent lipid peroxidation with-out inhibiting enzymes, such as CYP450 or its reductaseor by diverting electrons away from the peroxidativeprocess[26]. It was subsequently confirmed that bol-dine completely inhibited NAD(P)H/Fe3+–ATP-inducedhuman liver microsomal lipid peroxidation (Ki = 5�M)and CYP2E1 inactivation[26]. Microsomal lipid peroxi-dation caused the inactivation of numerous endoplasmicreticular enzymes including CYP450. Interestingly, bol-dine also effectively prevented the occurrence of lipidperoxidation in microsomal membranes incubated withCCl4 in the presence of NADPH, but failed to protect
ced
ent-
hylann-
ffectentsdlec-)H-by
nitedd-pro-itionted
[20] observed that some phenolic benzylisoquinoalkaloids inhibited Fe2+/cysteine-induced lipid peroidation (as TBARS production) of rat liver microsmal membranes. Later, using the same experimmodel, these investigators made similar observationa broader structural range of benzylisoquinoline aloids, including boldine and glaucine[21]. Subsequentlthe same authors found that some of these alkaalso protect deoxyribose against degradation inducFe3+–EDTA in the presence of hydrogen peroxide,sumably, acting as hydroxyl radical (HO•) scavenger[22]. In fact, in studies aimed at elucidating the mecnism of the antioxidant action of boldine, we establis
the CYP2E1 from undergoing the inactivation induby this halogenated hydrocarbon[26]. CCl4-mediatedinactivation of CYP2E1 may, therefore, be independof CCl4-mediated peroxidation. CCl4 may damage proteins via initially generated electrophilic trichloromet(CCl3•) radicals. Furthermore, CYP2E1 is indeedexcellent catalyst for CCl4 metabolism and inactivatioof CYP2E1 by CCl3• may well result from covalent binding to the protein, thereby escaping the protective eexerted by boldine on other free radical-mediated evtriggered by CCl3•. The ability of boldine is not limiteto the inhibition of the peroxidation dependent on etron transfer, as it also protects against the NAD(Pindependent oxidation of the membranes initiatedt-butyl-hydroperoxide (t-BOOH) decomposition[23].
Recently, Kubinova et al.[27] reported that wheadded to mouse liver microsomes, boldine inhibCYP1A-dependent 7-ethoxyresorufinO-de-ethylase anCYP3A-dependent testosterone 6-�-hydroxylase activities (enzymes which bioactivate promutagens andcarcinogens) and observed that 24 h after its addto a mouse hepatoma Hepa-1 cell line, it stimula
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 5
glutathioneS-transferase activity. These findings sug-gest the possibility that, in addition to its free radical-scavenging activity, boldine could also protect some vitalcell components not only by decreasing the metabolicactivation of potentially toxic xenobiotics but also byincreasing their removal. More studies are needed,however, to assess the actual chemoprotective poten-tial of boldine to act intracellularly through apparentlyantioxidant-independent mechanisms.
Although many substances have been found to beeffective inhibitors of microsomal lipid peroxidation,some features of the antioxidant activity of boldine maydistinguish it from other natural or synthetic antioxidantscommonly used in foods or as reference compounds. Infact, unlike agents, such as propyl gallate or phytic acid,which chelate iron and enhance its autoxidation, boldinewas shown to inhibit iron-mediated peroxidation at con-centrations 10 times lower than that of the iron catalyst[23].
In addition to lipids, proteins have been increasinglyrecognized as major biological targets of free radicals.In fact, their oxidative damage is regarded of impor-tance in the aging process and in the ethiogenesis ofcataracts, atherosclerosis and several inflammatory dis-eases. When enzymes are targeted, the oxidative mod-ification is generally evidenced as carbonyl group for-mation at the protein’s structure and is often associatedwith a loss of catalytic activity. Lysozyme is particu-larly, susceptible to peroxyl radical-mediated inactiva-tion. In early studies, we demonstrated the effectivenesso iva-tM selya ainstt oxylre lossp allyd for-m per-t idsb aseo n ofc lytica
overo hase rtiesi dsu deda tion.A ated
efficiency in preventing or retarding rancidity, their useas preservatives for human consumption has becomeincreasingly questioned on the basis of their alreadymentioned alleged toxicity[14,15]. Natural antioxidants,in turn, have enjoyed an increasing recognition and pop-ularity during the last decade. The industrial use ofnatural antioxidants has remained limited, however, pos-sibly due to their often higher cost and their relativelylower activity. Regarding the food preservative potentialof boldine, studies conducted by us revealed that it isparticularly efficient in protecting PUFA. Boldine wasfound to protect fish oil against spontaneous short- andlong-term O2-dependent thermal peroxidation, acting atconcentrations markedly lower than those required byother antioxidants. In fish oil, long-term (36 days, 25◦C)oxidation, boldine displays an activity similar to that ofquercetin and a two to three times greater activity than�-tocopherol, BHA or BHT[18]. Boldine also showed aremarkably greater potency than the latter tested antioxi-dants in protecting fish oil against Fe2+-induced rancidity[18]. More recently, it was also found to be more efficientthan quercetin and BHT in protecting bullfrog oil againstheat-induced oxidation[30] and stabilizingn-3-PUFA ofsardine oil against thermal oxidation with an efficiencygreater than that of BHA, BHT and�-tocopherol, butsimilar to that of quercetin[31].
Under most conditions alkoxyl- and peroxyl-free rad-icals play a crucial role in the initiation of heat- andmetal-induced oxidation of abiotic substrates. How-ever, in the presence of UV radiation, singlet oxygen
ffereles
d
onsathe
tere Ouctser
o
ct-ntokinge-
calsy or
f boldine at protecting this enzyme against inaction induced by AAPH-derived peroxyl radicals[17,28].ore recently, we showed that such effect is clossociated with a boldine-mediated protection ag
he oxidation of certain tryptophan residues by peradicals, which caused lysozyme inactivation[29]. Inter-stingly, the enzyme inactivation and tryptophanrevented by boldine were found to be mechanisticissociable from the AAPH-induced carbonyl groupation. Thus, it would seem that the antioxidant pro
ies of boldine comprise a protection, not only of liput also of protein targets, which, at least in the cf lysozyme, translates into preventing the oxidatioertain amino acids involved in the enzyme’s catactivity.
In addition to the apparent advantages of boldinether antioxidants in biological systems, this alkaloidmerged as also displaying potentially useful prope
n abiotic systems. Most oil- and fat-containing foondergo oxidative rancidity in the absence of adntioxidants, both during and after foodstuff preparalthough synthetic antioxidants have a demonstr
[O2(1�g)] is a major initiating species in foodstudeterioration. Boldine-containing boldo infusions wreported to effectively protect tryptophan molecufrom undergoing O2(1�g)-dependent oxidation[32].Subsequently, Zanocco et al.[33] found that boldine anglaucine were excellent quenchers of excited O2, partic-ularly, in polar solvents. Kinetic studies of the reactibetween these aporphines and O2(1�g) suggest thatcharge transfer complex be formed initially betweenaromatic ring and O2 and that such complex can laeither transfer the electron back, giving ground-stat2and the alkaloids or proceed to unidentified prod[33]. Thus, the ability of boldine (and possibly othaporphines) to quench O2(1�g) may also contribute tprotect foodstuffs against oxidative deterioration.
Intensive ultraviolet radiation, in addition to affeing foodstuff, can also promote harmful effects othe skin of humans. UV light may cause a mild sburn, skin cell DNA damage and premature skin aing possibly resulting in skin cancer. Since free radiare recognized to participate in either the aetiologthe development of most UV-induced skin lesions[34],
6 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
substances displaying free radical-scavenging capacityhave stemmed as potentially interesting photo-protectiveor photo-preventive agents. Interestingly, in addition toits antioxidant properties, the boldine molecule has twomajor absorption peaks, at 280 and 302 nm[9,10]. Thelatter would confer boldine a UV light-filtering propertyrelevant to a photo-protective action. In fact, Hidalgoet al. [35] showed boldine to be photo-unstable whenirradiated at wavelengths up to 300 nm and to display aphoto-protector effect against UV-B both in vitro andin vivo in mice. The photo-protection was evidencedthrough the prevention of the UV-induced increase inskin temperature of the rodents. More recently, Ran-can et al.[36] investigated the photo-filtering propertiesof boldine in humans. These authors observed that theapplication of boldine (25 mM) onto a 12 cm2 area of theback of volunteers protected their skin against erythemaformation to an extent slightly lower than that of a com-mercial sun cream for which a UV-protection factor of5 was informed. In the same study, it was observed thatthe in vitro irradiation of human T lymphocytes througha thin boldine-containing solution protected these cellsagainst loss of viability with a potency even greater thanthat shown by octylmethoxycinnamate, a UV-B refer-ence filter.
1.3. Antioxidant activity of boldine and ofstructurally related substances: structure–activityrelationships
ablediesidantted
)
ldinetiv-cetiony of
ofp intedule
roxylan
yedtion
of an ortho-semiquinone anion. However, the observa-tion that glaucine, theO-dimethylated boldine derivative,retains most of the parent molecule’s activity suggeststhat the presence of hydroxyl groups can add to, butis not essential for conserving the antioxidant activityof these aporphine analogues. Using the auto-oxidationof brain homogenates, Cassels et al.[28] confirmed theapomorphine molecule as the most active antioxidantamongst the tested aporphines, but the activity of boldineand glaucine suggests thatO-methylation causes only amarginal decrease of activity. A slightly lower activityof glaucine relative to boldine was reported by Ubeda etal. [22] using the Fe3+–EDTA–H2O2-induced deoxyri-bose degradation assay (hydroxyl radical-mediated).Glaucine had only half of the activity of boldine usingthe Fe2+/H2O2/ascorbic acid-induced liver microsomallipid peroxidation assay (hydroxyl radical-mediated) andthe AAPH-induced loss of lysozyme activity assay[37].The latter authors studied also the effect of halogena-tion (with bromo or iodine) of boldine at the C-3 posi-tion (Table 1) and observed that such modification doesnot affect lyzozyme protection, but it almost double theantioxidant effectiveness on the microsomal lipid perox-idation assay. The results seen in the latter assay are inline with the recognized importance of the lipophilicityin chain-breaking lipid peroxidation within membranes.
On the early structure–activity research by Martınezet al. [21], it was suggested that, in the absenceof phenolic groups, aporphines could be easily oxi-dized to dehydro- (actually 6a,7-didehydro) and oxo-
nes.da–Hckulddi-ion.t
, isiv-tsrated, theuldgingon-e ofemsringity.
After the initial observations by Rıos et al. [20]that some phenolic benzylisoquinolines could beto interfere peroxidative processes, several stuaddressed the free radical-scavenging and/or antioxproperties of a larger number of isoquinoline-relaalkaloids. Using a chemically induced (Fe2+–cysteinemicrosomal lipid peroxidation assay, Martınez et al.[21]found that, amongst several aporphines tested, bo(IC50 = 20�M) displayed almost twice as much acity as isoboldine (Table 1), suggesting the importanof the presence of a hydroxyl group in the C1 posiof aporphine structures. Similarly, the greater activitthe aporphine bulbocapnine relative to anonaine (IC50 of12.5�M versus 27�M) could be explained in termsthe importance of the presence of a hydroxyl grouposition the C11 (Table 1). The same authors reporthat the dihydroxy-containing apomorphine molec(Table 1) exhibited the highest activity (IC50 = 3.3�M),suggesting that the presence of a catechol with hydgroups at the C10 and C11 positions would beimportant feature towards the high activity displaby most aporphines, possibly endowing the forma
(in fact, 6-nor-7-oxo-4,5,6,6a-tetradehydro) aporphiThus, based on the latter, Cassels et al.[28] suggestethat in non-phenolic aporphines, the benzylic C–6bond may be the initial point of free radical attaand that the neighbouring nitrogen lone pair wostrongly contribute to the stabilization of the intermeate radical by further extending electron delocalizatSupporting the latter, Cassels et al.[28] showed thacompared to glaucine,N-methylglaucinium (Table 1),an analogue in which the nitrogen is protonatedcompletely devoid of activity (in the lysozyme actity protection assay). Milian et al.[38] reported thaN-methylboldinium (Table 1) is only half as active aboldine in scavenging reactive oxygen species geneby a hypoxanthine–xanthine oxidase system. Thusstabilization provided by the nitrogen lone pair wobe essential for conserving the free radical-scavenactivity seen in both, the phenolic as well as in the nphenolic aporphines. With regard to the importancthe piperidine ring in the aporphine structure, it sethat, at least for boldine, the presence of suchwould not be essential for manifesting such activ
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 7
In fact, N-allylsecoboldine (Table 3), a phenanthrenicstructural analogue of boldine in which the nitrogenatom is part of a side chain (N-allyl derivated), hasbeen reported to exhibit interesting antioxidants proper-ties[39]. Like boldine[17,23], N-allylsecoboldine scav-enges hydroxyl and peroxyl radicals and efficiently, pre-vents lipid peroxidation in rat brain homogenates anderythrocyte membranes. Although, there are no stud-ies aimed at addressing the relative activity of boldineandN-allylsecoboldine, Milian et al.[38] compared theability of aporphines with that of the correspondingequivalents with a phenanthrene skeleton for scaveng-ing ROS generated by hypoxanthine–xanthine oxidase.These investigators found that boldine is less active thanany of the tested phenolic (C3–OH and C7–OH) phenan-threne analogues: secoboldine,N-methylsecoboldineand N-methylsecoboldinium (N-dimethylsecoboldine)(Table 3). The higher activity reported for phenolic-phenanthrenic relative to phenolic-aporphinic moleculesmay reside on the existence of a third benzylic ringwhich may confer a greater capacity to delocalize thephenoxy free radical to the former. Nonetheless, thesame group observed that boldine was still more activethan the non-phenolic phenanthrene,O-dimethylated-N-methylsecoboldinium (Table 3) and that the presenceof free HO groups was an absolute requirement for theactivity of phenanthrene derivatives.
On the other hand, in structure–activity studies con-ducted on benzylisoquinoline alkaloids, Martınez et al.[21] found that reticuline and laudanosine (analogueso vely)( u-d ndt en-z atO en-z orec ctivem ben-
zylisoquinolines. Noteworthy, laudanosoline, despitepresenting two cathecols, shows only half the activity ofapormorphine, which exhibits only one of such moieties.About the latter, Cassels et al.[28] postulated that thesuperior ability of aporphines to trap free radicals is asso-ciated with an increased spin delocalization of phenoxyradicals in the biphenyl system. The resulting benzylicfree radicals, by analogy with phenoxy radicals, wouldpresumably be better stabilized in aporphines, as com-pared to their benzylisoquinolines-derived counterparts,by extended conjugation across the aporphine biphenylsystem.
1.4. Pharmacological properties of boldine andrelated substances associated with their antioxidantactivity
Here we review evidence showing the ability of bol-dine to limit experimentally induced processes relevantto pathophysiological conditions in which free radicalsand/or peroxidation products are involved as intracellu-lar damage mediators.
Compelling in vitro and in vivo evidence impli-cates free radicals as major initiators and/or mediatorsof biochemical events leading to cell damage[13,40].Stemming from the free radical-scavenging ability ofboldine, the pursuit of its pharmacological potentialhas included establishing its effectiveness in protect-ing cells against oxidative and lytic damage. Erythro-cytes are particularly susceptible to undergo oxidative
iso-rane
otethege ofon,lysisdage
f the aporphines isoboldine and glaucine, respectiTable 2), showed very poor activity. However, laanosoline (Table 2), a bicathecol molecule, was fou
o exhibit the greatest activity amongst all tested bylisoquinolines (IC50 = 6.8�M). Thus, it appears th-methylation substantially affects the activity of bylisoquinolines. The presence of at least one or matechols is a common feature amongst the most aolecules within both, the the aporphines and the
stress. AAPH has been shown to induce in intactlated erythrocytes extensive peroxidation of membcomponents (e.g., lipids and proteins) and to promstructural alterations, which lead to the rupturingerythrocyte membrane and to the subsequent leakahemoglobin[41]. Using a rat erythrocyte suspensiwe showed that boldine cytoprotects against cellinduced by AAPH[42]. Boldine prevented, time- anconcentration-dependently, the AAPH-induced leak
8 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
of hemoglobin into the medium. Such cytoprotectionwas observed whether the antioxidant was added to theerythrocyte suspension 1 h prior to or simultaneouslywith the azo-compound, suggesting that boldine was nei-ther metabolized nor oxidized during such preincubationperiod[42].
The membrane permeability transition of mitochon-dria is known as a central event in the course of a varietyof toxic and oxidative forms of cell injury as well asapoptosis induced by xenobiotics or drugs[43]. Open-ing of the mitochondrial permeability transition porehas been shown to induce depolarization of the trans-membrane potential, release of small solutes, release ofCa2+ and cytochromec, osmotic swelling and loss ofoxidative phosphorylation. The oxidation of dopamine(DA) and 6-hydroxy-dopamine (6-OH-DA) in neuronsinduces free radical formation, inhibition of mitochon-drial respiratory chain and opening of the mitochondrialtransition pore, which is inhibited by oxidant scavengers.Based on the latter, Youn et al.[25] examined the pro-tective effect of boldine on catecholamine (DA and 6-OHDA)-induced brain mitochondria dysfunction and onPC12cell death. At concentrations (10–100�M), whichdid not affect mitochondrial permeability, boldine sub-stantially decreased the effect of catecholamine (CA) onmitochondria swelling and attenuated the alteration ofmitochondrial membrane potential. At the same concen-trations, boldine attenuated the CA-induced decrease inthioredoxin reductase activity and increase in thiol oxi-dation and at 100�M also decreased the mitochondrial
s,
essedthors
ithagecedical-s byro-
tion./kg,inst,6-dbolite
es aincestri-
tio
in the control mice[37], its failure to promote neuropro-tection against MPTP reflects that the concentration ofboldine at the striatum might be enough to modify cate-cholamine metabolism but not to counteract the oxidativestringency induced by MPP+ at such site. Future studiesaimed at assessing the neuroprotective potential of bol-dine should establish its property to react with MPP+ inrelevant in vitro models and should address in vivo theability of boldine to cross the blood–brain barrier.
Early work by Lanhers et al.[44] showed that micro-molar concentrations of boldine protected isolated rathepatocytes against damage induced byt-BOOH. Sub-sequently, they showed that lower concentrations ofboldine was highly effective in protecting rat liver micro-somal membranes against lipid peroxidation inducedby t-BOOH [23]. Boldine, however, did not preventthe early (within 60 s) and sudden decline (by 50%) ofreduced glutathione (GSH) and the equivalent increasein oxidized glutathione (GSSG)[45] attributed to a GSH-peroxidase-catalyzed reaction witht-BOOH rather thana non-enzymatic reaction witht-BOOH-derived free rad-icals. In fact, the levels of both GSH and GSSG recoveredto near-basal levels after 20 min, either in the presence orabsence of boldine, indicating that the protective actionsof boldine do not depend on the prevention of changesin GSH/GSSG concentrations. On the other hand, thedelayed addition of boldine (10 or 20 min after) to cellsincubated witht-BOOH, while effectively blocking anyfurther increase in TBARS, totally failed to prevent thesubsequent peroxide-induced loss of cell viability, and
hus,tionntin-atedcu-
, attra-asuentionguchldineapstes
nstoxicen-alu-hers
release of cytochromec. Consistent with such actionYoun et al.[25] demonstrated that boldine (10–100�M)decreased DA-induced death in PC12 cells (as assby the MTT assay and caspase-3 activity). The au[25] concluded that, since boldine effectively reacts wHO• radicals, its ability to attenuate CA-induced damof brain mitochondria and to decrease the DA-induapoptosis and cell death would reside on its radscavenging properties. While the in vitro observationYoun et al.[25] suggest a potential for boldine as neuprotective agent, subsequent work by Loghin et al.[37]conducted in vivo offered no support for such contenThus, when injected (s.c.) in mice at a dose of 40 mgboldine was found to be ineffective in protecting agathe neurotoxic action of 1-methyl-4-phenyl-1,2,3tetrahydropyridine (MPTP)[37]. The latter compoungenerates in dopaminergic neurons an active meta(1-methyl-4-phenyl-pyridinium, MPP+), which inhibitsthe mitochondrial electron transport chain and inducsyndrome closely resembling Parkinson’s disease. Sat the same dosing, boldine was effective to increaseatal levels of DOPAC and HVA and the HVA/DA ra
in fact, was associated with increased cell death. Tit is possible that the early increase in lipid peroxidaproducts is sufficient to either signal or assure the couance of other events, whether mechanistically relor not, leading to cell death. Conversely, the preinbation of hepatocytes with boldine during 150 minwhich time no boldine could be detected either inor extracellularly, prevented lipid peroxidation and was effective at protecting cells against the subseqaddition of t-BOOH as was the simultaneous additof boldine andt-BOOH to cells preincubated durin150 min under control conditions. On the basis of sobservations, we suggest that early exposure to bomay trigger events resulting in cytoprotection, perhinvolving the formation of boldine-derived metaboliwith similar antioxidant properties[45].
In addition to t-BOOH, halogenated hydrocarbohave been widely used as paradigmatic hepatoagents[46]. Using carbon tetrachloride as experimtal hepatotoxin in vivo, boldine has also been evated as a cytoprotectant. Work conducted by Lanet al.[44] showed that boldine prevented CCl4-induced
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 9
hepatitis in mice. The latter authors reported that bol-dine, given i.p. at a dose of 10 mg/kg partially protectedmice against hepatotoxicity (as assessed by an increasein plasma GPT) when administered 30 min before thehalocarbon. The fact that the protection afforded by bol-dine was only partial (less than 50%) may be interpretedas an indication that the dose of boldine used by Lan-hers et al.[44] was insufficient, even though the dose ofCCl4 used in this study (0.03 mL/kg) was far below thatemployed by other investigators[46]. The partial protec-tive effect of boldine may relate to the in vitro observationby Kringstein and Cederbaum[26] that boldine protectsliver microsomal membranes against CCl4-induced lipidperoxidation but fails to protect against CYP2E1 destruc-tion. Presumably, before the cytochrome inactivation iscompleted, CCI3• radicals under formation binds cova-lently to an array of macromolecules whose functions aremetabolically key in defining the hepatocyte integrity.In the latter framework, a partial degree of cytoprotec-tion would be expected. On the other hand, it is alsonecessary to consider the fact that for an antioxidant toprotect against tissue damage induced by a (peroxidiz-ing) xenobiotic it is necessary that both substances (orthose underlying their action) should be present simul-taneously, for enough time and at the same cellular andsubcellular locus. It is also a requirement that the antiox-idant be present at its site of action in concentrationsrelevant to its cyto-protective action. Such conditionscan often be met in vitro, but achieving them in vivois much more complex. Pharmacokinetic studies con-d gestt pato-p d toh os-sH ncer omo maw r toe thefi ent,p noti ellu-l d, itw danta w bol-d g tol ivoe ess.
andl acidi eac-
tive species are also produced in substantial amountsduring inflammatory phenomena in association withleucocytes infiltration[40]. Consequently, antioxidantmolecules, which may interfere with ROS generationmay also display anti-inflammatory properties. Usingthe carrageenan-induced paw oedema assay, Lanherset al. [44] showed that the administration of a puri-fied boldo leaf extract to rats had an anti-inflammatoryeffect that they could not reproduce when boldine wasadministered (10 or 20 mg/kg; i.p.) instead. Subsequentwork by us [47], however, demonstrated that boldineis very effective (ED50 = 34 mg/kg; p.o.) in reducingcarrageenan-induced paw oedema in guinea pigs and inpreventing (60 mg/kg; p.o.) the increase in rectal tem-perature seen in rabbits treated with bacterial pyrogen.These anti-inflammatory and antipyretic effects of bol-dine were mechanistically supported by the observationthat upon its addition to a rat aortal ring preparation, bol-dine effectively inhibited prostaglandin biosynthesis (as6-keto-PGF). The exact mechanisms by which boldineinhibits PG synthesis and the relationship between suchaction and its anti-inflammatory and antipyretic effectsremains to be investigated.
On the other hand, enhanced generation of reactiveoxygen and nitrogen metabolites are also implicated intissue injury observed in several inflammatory boweldiseases (IBD). Thus, antioxidants could function ame-liorating and/or preventing the inflammation and cyto-toxicity seen in experimental IBD models. Based on theantioxidant and anti-inflammatory properties of boldine
ticn ofivenminandopictedrox-d thesed
hlyNs
nti-deldi-ellsOS
of theider
ucted by us on tissue boldine concentrations sughat for the doses used in the above-referred herotection experiments, boldine might be expecteave attain potentially effective antioxidative (and pibly cyto-protective) concentrations in the liver[11].owever, it must be kept in mind that the cleara
ate of boldine from plasma is extremely high. Frur kinetic studies, the half-life of boldine in plasas estimated to be around 30 min. Thus, in ordensure the continued presence of boldine duringrst 8–12 h post-administration of the hepatotoxic ageriod during which the serum enzyme levels had
ncreased significantly but some damaging intracar events had been presumably already triggereould probably be necessary to have the antioxidministered repeatedly. Future assessments of neine analogues—with structural modifications leadin
onger half-lives—might merit a more detailed in vvaluation of their potential hepatoprotective usefuln
ROS are involved in the cycloxygenase-ipoxygenase-mediated conversion of arachidonicnto pro-inflammatory intermediates and that these r
[44,47] Gotteland et al.[48] studied in vivo the effecof boldine in a rat model of IBD in which the colondamage was induced by the intrarectal administratioacetic acid. These investigators found that boldine, gas a single dose (100 mg/kg) intrarectally and 30before acetic acid, prevented oedema formationafforded substantial protection against the macroscand histologic injury. The mucosa of boldine-treaanimals exhibited a substantial reduction in myelopeidase activity, suggesting that the antioxidant lowereextent of neutrophils infiltration into the acid-expotissue. Noteworthy, recently Milian et al.[38] reportedthat low micromolar concentration of boldine are higeffective in preventing ROS production by human PMinduced by the bacterial chemotactic peptide,N-formyl-methionylleucyl-phenylalanine (f-MLP). Thus, the ainflammatory effect of boldine seen in the IBD mo[48], could relate to antioxidant actions that work, inrectly via lowering the presence of ROS-producing cin the affected tissue and directly, via decreasing Rproduction by the same cells. Future assessmentspotential usefulness of boldine in IBD, should cons
10 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
other experimental models of chronic inflammation andusing also other routes of boldine administration.
Evidence supporting the hypothesis that an increasedoxidative could be involved in the pathogenesis and pro-gression of diabetic tissue damage has prompted theexperimental and clinical evaluation of the potentialof antioxidants in the prevention and/or treatment ofdiabetes. In the single study available addressing theusefulness of boldine as anti-diabetic agent, Jang et al.[24] investigated the effect of boldine in preventing thedamage induced by streptozotocin (STZ) in rats. Theadministration of STZ to rats induces damage to pancre-atic beta cells and results in diabetes by mechanisms thatare not clearly understood, but that involve an increasedformation of hydroxyl radicals and other ROS in associa-tion with the prior generation and/or subsequent decay ofhighly reactive STZ carbonium radicals[49]. Oxidativealterations within the mitochondria, believed to con-tribute to the dysfunctions seen in diabetes, are partiallyreplicated in the STZ-induced diabetes model[49]. Asreported by Jang et al.[24], boldine (given at a dose of100 mg/(kg day) in drinking water during 8 weeks) sig-nificantly prevented the elevation of blood glucose levelsand the weight loss induced by STZ and attenuated theelevation of carbonyls and malondialdehyde levels seenin pancreas mitochondria. The antioxidant effects of bol-dine, were not evident, however, for carbonyls in livermitochondria nor for malondialdehyde in mitochondriafrom kidney. Interestingly, Jang et al.[24] observedthat in vitro boldine inhibited ROS production by iso-
cinedty inals.g theateddinent of
tion,the
mheon-iz-nd
edan-er-ma
obtained from LDL-deficient mice fed simultaneously ahigh fat atherogenic diet and boldine (p.o. 1 or 5 mg/day)during 12 weeks. Whole plasma lipids isolated fromboldine-fed (LDLR−/−) mice were considerably lesssusceptible to in vitro oxidation compared to that ofcontrol animals. On the same in vivo atherogenic studyprotocol, boldine was shown to significantly decreasethe areas of the aortic atherosclerotic lesions, by 22 and44% for the 1 and 5 mg/day 12-week dosing, respec-tively. The study by Santanam et al.[50] shows thatrepeatedly administered, boldine effectively protectsLDLR−/− mice from developing atherosclerotic lesions.Since boldine exerted such effect without altering plasmacholesterol, triglycerides, LDL and HDL levels, its anti-atherogenic effects may well be attributed to its antioxi-dant properties.
Platelet activation seen in hypercholesterolemic indi-viduals is often associated with the synthesis of pro-inflammatory cytokines and with the release of a myriadof chemokines, which contribute to the development andprogression of atherosclerotic plaques. In the contextof testing the potential antiplatelet usefulness of apor-phine antioxidants, Teng et al.[51] found that in plateletobtained from rabbits, boldine inhibited the aggregationinduced by arachidonic acid and collagen, but not thatinduced by platelet-activating factor (PAF), thrombin ora thromboxane analogue U46619. They concluded thatthe antiplatelet effect of boldine mainly resulted fromits inhibition of thromboxane A2 formation from arachi-donic acid. It remains to be established whether this in
eer-
ec-a
heelln-
2-sso-hinsionc-pro-
einorsPAown-an-ine
lated liver mitochondria when treated with antimyc, a respiratory inhibitor. In vivo boldine normalisthe elevated Mn–SOD and GSH-peroxidase activimitochondria of the pancreas of STZ-treated animThese results suggest that boldine, by suppressinoxidative stress existing in the pancreas of STZ-treanimals, opens up the exploration of the use of boland its antioxidant-related analogues in the treatmediabetes-associated free radical overproduction.
Atherosclerosis involves three processes, oxidainflammation and hypercholesterolemia. In view ofpreviously established antioxidant[17,23,28,42]andanti-inflammatory[47] properties of boldine, Santanaet al. [50] studied the in vitro effect of boldine on toxidizability of human LDL (assessed by diene cjugate formation) and its ex vivo effect on the oxidability of whole plasma obtained from boldine-fed acontrol LDL receptor-knockout (LDLR−/−) mice. Invitro, boldine inhibited human LDL oxidation inducby copper (5�M) in a concentration-dependent mner (0.5–2.5�M). The authors also investigated coppinduced diene conjugate formation ex vivo in plas
vitro inhibitory platelet aggregation ability of boldinalso contributed to its in vivo anti-atherogenic propties recently shown by Santanam et al.[50].
Oxidative stress is also implicated in the molular mechanism of tumour promotion by inducingdown-regulation of cellular gap junctions (GJIC). Tlatter structures play a key role in the control of cgrowth, development and differentiation. The dowregulation of GJIC induced by substances, such as 1O-tetradecanoylphorbol-13-acetate (TPA), occurs in aciation with an increased production of oxidants witcells. Since the latter can lead to a rapid clonal expanof initiated cells, in vitro models of gap junctional funtion can be used to screen the potential anti-tumourmoting properties of an antioxidant. Hu et al.[52] testedthe efficacies of boldine and itsO-dimethylated glaucinto inhibit the TPA-induced down-regulation of GJICWB-F344 rat liver epithelial cells. These investigatshowed that the co-incubation of these cells with Tin the presence of these aporphines prevented the dregulation of GJIC in a concentration-dependent mner. At 50�M concentration each, boldine and glauc
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 11
completely restored the gap junction. Interestingly, theseantioxidants prevented TPA-induced increment in thecellular oxidative tone (measured using the dichloroflu-orescin probe) and inhibited TPA-induced translocationof protein kinase C (PKC) into the membrane. The lat-ter effects took place concomitantly with an inhibitionof TPA-induced PKC-dependent hyperphosphorylationand subsequent internalization of gap junctional con-nexin 43 (cx43) into the plasma membrane. These resultssuggest that boldine and its analogue glaucine possessesanti-tumour promoting activity through their ability tointerfere in TPA-induced down-regulation of GJIC.
1.5. Pharmacological properties of boldine whichare not necessarily associated with its antioxidantactivity
Chagas’ disease (American trypanosomiasis) iscaused by several strains ofTrypanosoma cruzi and rep-resent a permanent threat to almost 20% of the popula-tion of Latin America. Due to the toxicity of the syntheticdrugs used to treat this disease (e.g., nifurtimox andbenznidazole), some natural products may be seen as apossible safer alternative. Based on early work showingthat several synthetic antioxidants inhibited the respira-tion and growth ofT. cruzi, Morello et al.[53] lookedfor the trypanosidal activity of the aporphines boldine,glaucine and apomorphine. The investigators found thatall three compounds completely inhibited the growth ofepimastigotes of the Tulahuen strain, LQ strain and DM28 erc erea fcs owtho s-pu inea gas’d
ation. The very high boldine concentration employedby the latter investigators limits, however, any inter-pretation of their results. Using notably lower concen-trations (10–30�M), earlier work by Gonzalez-Cabelloet al. [55] suggested that boldine could exert in vitroan immumodulating effect on natural killer cells frompatients presenting low activity and decreased blastoge-nesis in both, normal individuals and in patients withchronic lymphocytic leukemia. Finally, in a latter workby Philipov et al.[56], it was found that boldine inhibitthe in vitro concanavalin A-induced proliferation ofmouse splenocytes. Although these works suggest somepossible immunomodulating properties of boldine, theactual potential of this aporphine to favourably modifycellular immune functions require confirmation and fur-ther assessment.
Several studies addressing the effects of boldine onthe smooth muscle have pointed out the possibility thatthis alkaloid may exert some muscle-relaxing effects,which could be of potential pharmacological interest.Thus, boldine was early shown to inhibit the intesti-nal smooth muscle activity in the anaesthetized cat[57]and to prolong the intestinal transit in mice[58]. Work-ing in a rat ileum preparation, our laboratory observedlater that boldine exerts a concentration-dependent relax-ation effect by directly interfering with the cholinergicmechanism associated with the contraction[59]. Relatedto the latter, we also observed that the administrationof a boldine-containing boldo-dried extract to healthyvolunteers significantly prolongs the oro-cecal intestinal
l-oldoded
s ofara-of
,wasd in
eice,liaryoidd ofhave
ineuslyf rat
n an, by
8c clone ofT. cruzi at 500�M (with IC50 ranging from0 to 120�M, for all strains). Although much highoncentration was required (1 mM), these alkaloids wlso shown to inhibit (by 25–30%) the respiration oT.ruzi epimastigotes in all strains. Morello et al.[53] haveuggested that these aporphines might inhibit the grf T. cruzi by inhibiting mitochondrial electron tranort. Although the work by Morello et al.[53] openedp the possibility of exploring the usefulness of boldnd some of its analogues for the treatment of Chaisease.
Macrophages play an important role in host defeechanisms. Their principal functions includehagocytosis of foreign particles, showing the pnce of antigens and the production of cytokinesadical species. A variety of plant-derived materan stimulate the immune system. Moreira et al.[54],ho investigated the effect of boldine on mouse p
oneal macrophage functions by the liberation of H2O2,ound that boldine (6 mM) presented a very low mlatory activity on the immune system, so the alkaery poorly enhanced the macrophage peroxide l
transit rate[6]. The extent to which, in addition to bodine, some of the abundant flavonoids present in bmay have contributed to this effect cannot be excluyet. In turn, the cholagogic and/or choleretic effect(boldine-containing) boldo extracts and other preptions [1], seem to also occur upon administrationpure boldine to relevant animal models[60–62]. Thuswhen given by gavage to dogs (2 mg/10 kg), boldinereported to almost double the total content of solibile, without modifying the biliary volume[60]. How-ever, Levy-Appert-Collin and Levy[61] reported that thdirect application of boldine into the duodenum of mat a dose of 50 or 100 mg/kg, incremented the biflux, by 15 and 60%, respectively. Although the flavonglycosides of boldo have been found to be devoidecholeretic activity by themselves, these compoundsbeen claimed to enhance the effect of the alkaloids[61].
In addition to an intestinal-relaxing action, boldhas also been reported to induce relaxation in previocontracted (acetylcholine-induced) smooth muscle outerus[63] and to block the neuromuscular action imuscle phrenic-nerve mouse diaphragm preparatio
12 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
direct interaction with the post-synaptic acetylcholinereceptor[64]. The muscle-relaxing effect of boldine(IC50 of 13.5�M) was found to be reversible andconcentration-dependent and was suggested to be mostlyan action on nicotinic ACh receptors[64]. Hue et al.[65], working on the cockroach ganglion, observed thatboldine acted as a specific nicotinic antagonist of theinsect CNS, without displaying effects on muscarinicand GABAergic receptors. Consistent with the latter,Chulia et al.[66] showed that boldine antagonized the�1-adrenoceptor in the guinea pig aorta, but had noeffect on acetylcholine-induced contraction of the tra-chea. Thus, data suggest that boldine had higher speci-ficity towards the nicotinic compared to the muscarinicreceptor.
In addition to the its above described actions, boldinehas been shown to block Ca2+ channels in rat uterus[63], rat aorta[67] and rat cerebral cortex[68], possi-bly through the benzothiazepine receptor site, but withan affinity considerably lower (between 1 and 2 ordersof magnitude) than diltiazem. More recently, Eltze etal. [69] confirmed the Ca2+ channel antagonist propertyof boldine in a rat perfused kidney preparation, show-ing that boldine effectively blocks the vasoconstrictioncaused by elevated extracellular potassium. When testedon mouse diaphragm and in isolated sarcoplasmic retic-ulum membrane vesicles, boldine was found to act byinducing the release of Ca2+ from internal Ca2+ storagesites of skeletal muscle[70]. Besides its Ca2+ channelblocking activity, boldine was also able to antagonize
a
ratfoldA
ep-etelythe
ol-therlays
aticlsothe
-like,may
inedi-fini-
ties for the D1- and D2-like receptors[73,74]. How-ever, in vivo (i.p. 40 mg/kg), boldine was unable todisplace [3H]-raclopride, a selective D2-ligand, fromeither mice striatium or olfactory bulbs; boldine onlymarginally displaced [3H]-SCH23390, a selective D1-ligand, from mice striatum[73]. Given at the same dose,boldine did not modify the apormorphine-elicited climb-ing, sniffing and grooming behaviours. However, in theapomorphine-induced rat yawning and penile erectionmodel (associated with D2-receptor), boldine inhibitedboth behaviours by more than 50% but did not affectstriatum dopamine metabolism[73]. Therefore, althoughboldine may act as a dopamine antagonist, its apparentlypoor access to at least to certain regions of the CNS[73],added to its very short plasma half-life[11], do not allowthis property to be easily revealed in some in vivo exper-iments.
1.6. Toxicological studies and safety concerns onboldine
The precedent of the prolonged tradition of pharma-ceutical use of boldine and boldine-containing boldopreparations suggests that boldine exhibits low toxic-ity. In fact, relatively high doses are needed to induceside effects, toxicity or lethality in several mammalianspecies. Early studies by Kreitmair[62] reported that500 and 1000 mg/kg (p.o.) were required to induce thedeath of mice and guinea pigs, respectively. The latterauthors showed that considerably lower doses, 250 and
ice(i.v.)
i-o die
ro-ith-to
esce
ofngesto
to
cy,ra-d no
�1-adrenoceptors in rat aorta[67], in guinea pig aort[66] and in rat cerebral cortex preparations[68]. Bindingcompetition studies conducted with [3H]-prazosin incortex, indicated that boldine exhibited around 65-greater affinity for the native (high affinity) subtypereceptor (pKi = 8.31) compared to the subtype B rector (pKi = 6.50)[71]. Consistent with the latter, Eltzeal. [69] found that boldine has an affinity approximat25- and 15-fold higher for the subtype A compared tosubtypes B and D�-adrenoceptors, respectively. Bdine was also found not to discriminate between�1-adrenoceptor subtypes B and D and�2-adrenoceptosubtypes A–C, at which the drug consistently dispmicromolar affinity.
As a close structural congener of the paradigmdopaminergic agonist (R)-apomorphine, boldine has abeen reported to exert some inhibitory effects atcentral nervous system. These include neurolepticanticonvulsivant and antinociceptive actions, whichprobably be mediated through the blocking of dopamD2 receptors[72]. Subsequent in vivo studies incate that boldine presents in vitro good binding af
50 mg/kg (i.v.), were required to induce the death of mand guinea pigs, respectively, whereas 25 mg/kgwere required to induce the death of cats[62]. Studiesconducted later by Levy-Appert-Collin and Levy [61],estimated an LD50 of 250 mg/kg (i.p.) in mice. Most anmals employed in the above studies were reported tby respiratory failure.
Studies conducted by Moreno et al.[75] reportedthat boldine has no mutagenicity in the SOS chmotest and in several Ames tester strains, with or wout prior metabolic activation. Boldine was not ableinduce point and frameshift mutations in haploidSac-charomyces cerevisiae cells[75]. Subsequently, Tavarand Takahashi[76] reported that boldine did not indua statistically significant increase in the frequencychromosome aberrations or sister chromatid exchain vitro in human peripheral blood lymphocytes (up40�g/mL) or in vivo, in mouse bone narrow cells (up900 mg/kg, administrated p.o.).
Regarding the toxicity of boldine in pregnanAlmeida et al.[77] observed that its acute administtion to rats during their early pregnancy phase induce
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 13
foetus resorption and no foetal malformation when givenp.o. at 500 mg/kg. A weak but significant abortive andteratogenic effect was evident, however, at 800 mg/kg.Studying the effects of long-term administration, thesame authors observed a low degree of hepatotoxicity,assessed by blood transaminases or urea levels, in ratsgiven boldine p.o. daily at 800 mg/kg for 30 and 60days but not seen at 500 mg/kg. No hepatic histologi-cal modifications were observed at a dose of 800 mg/kgadministered for 90 days[77].
Recently, a case of anaphylactic reaction alter theintake of a boldo infusion was reported in a 30-year-oldman with personal history of allergic rhino-conjunctivitis[78]. The authors suggest that a type I IgE-mediatedimmunologic mechanism as responsible for the patient’sanaphylactic symptoms. It is not clear, however, whethersuch a reaction can be attributed to boldine. In fact, IgE
symptoms have also been observed with other boldine-free infusions like coffee, cacao and tea. Finally, alsoin a recently reported case, a several fold increase inblood transaminases was detected in an elderly malepatient with fatty liver who was taking daily herballaxatives, which contained boldo leaf extracts. Transam-inases returned to normal following withdrawal of thelaxative[79].
2. Final remarks and future research needs
Research conducted mostly over the last decade,clearly substantiates the ability of boldine to act as apotent free radical-scavenger and antioxidant molecule.As reviewed here, ample evidence shows that bol-dine can react with high efficiency towards a broadscope of reactive species to either prevent or retard
Table 4Summary of studies on the free radical-scavenging and antioxidant properties of boldine
14 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
the targeted oxidation of lipids, proteins and nucleicacids. Structure–activity studies have provided valuableinsights into the main structural features underlyingthe antioxidant properties of boldine. Boldine has beenbroadly shown to exert potent cyto-protective effects inmodels of oxidative stress-induced damage. In vitro, forinstance, it was able to protect isolated red blood cells,hepatocytes and neurons from undergoing cell lysis. Invivo, boldine has been found to prevent or largely ame-liorate the oxidative damage and the cell injury inducedto the pancreas and to the colon epithelium by oxidantsin rat models of diabetes and ulcerative colitis, respec-tively. A large number of experimental studies haveproven the effectiveness of boldine in preventing variousoxidative stress-related pharmacological effects includ-ing anti-inflammatory, antipyretic, anti-tumour promot-ing, anti-platelet and anti-atherogenic effects. In viewof the increasing recognition of the participation of freeradical-mediated oxidative events in the ethiogenesis of
various cardiovascular, tumoural, inflammatory and neu-rodegenerative pathologies, it would seem that, at thispoint in time, the major prospects of pharmacologicalapplication of boldine would stem from its ability toprevent oxidative stress. However, as reviewed here, anumber of additional studies have shown that boldine canalso promote some pharmacological effects which do notappear to be associated with its antioxidant properties,to mention are the muscle-relaxing, choleretic and/orcholagogic, anti-trypanocidal and immuno- and neuro-modulating effects of boldine. Future studies on the latteractions of boldine could well broaden its potential phar-macological scope. Data available on its toxicity, whichhas been conducted mostly in isolated cells and in variousanimal models, points to a relatively low toxicity of bol-dine. However, its actual innocuousness in humans stillremains to be established. A major limitation to assessthe latter, however, would reside on the current lack ofinformation on the actual doses of boldine that would be
Table 5Summary of pharmacological actions of boldine associated with its antioxidant activity
Pharmacological action Model Effect References
Anti-inflammatory Carrageenan-induced paw oedema in guinea pigs.Oral administration of boldine
Effective reduction of oedema [47]
Acetic acid-induced colonic damage in rats.Intrarectal administration of boldine
Prevention of oedema, tissue damage andneutrophil infiltration
[47]
f-MLP-induced ROS production by isolated humanPMNs
In vitro prevention of ROS production byhuman PMNs
[38]
mia in
damag
tro
in LDL−
by ara
IC in W
Isolated rat aorta rings
Antipyretic Bacterial pyrogen-induced hypertherOral administration of boldine
Antidiabetic STZ-induced pancreatic (beta cells)Oral administration of boldine
Antiatherogenic Copper-induced LDL oxidation in vi
Diet-induced atherosclerotic damagemice. Oral administration of boldine
Antiplatelet Platelet aggregation induced in vitroacid and collagen
Anti-tumour promoting TPA-induced down-regulation of GJrat liver epithelial cells
Photo-protection UV-irradiated mice skinPhoto-protection UV-irradiated human back skin
Hepatoprotection CCl4-induced hepatotoxicity in rats. Intraper
administration of boldine
In vitro inhibition of prostaglandinbiosynthesis (6-keto-PGF)
[47]
rabbits. Prevention of hyperthermia [47]
e in rats. Prevention of hyperglycaemia andweight loss
[24]
Reduction of TBARS and carbonyllevels in various tissuesNormalization of mitochondrialantioxidant enzymes
Inhibition of the oxidation of isolatedhuman LDL
[50]
R/− Inhibition of ex vivo plasma oxidationand reduction of atherosclerotic aortalesions
[50]
chidonic Inhibition of aggregation [25]
B-F344 Inhibition of TPA-induceddown-regulation
[52]
Prevention of TPA-induced increase incellular oxidative tone and in PKCtranslocation
Prevention of skin rise temperature [35]Prevention of skin erythema formation [36]
itoneal Prevention of liver damage (plasma GPT) [44]
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 15
required to exert in humans some of the pharmacolog-ical actions reported so far and demonstrated solely inanimals.
The effectiveness of boldine as an antioxidant and itsrelatively low toxicity observed in animals now justifythe pursuit of studies aimed to explore its actual therapeu-tic value in phase 1 clinical studies including assessingits pharmacokinetics and major biotransformation path-ways. Such studies, to be conducted in healthy subjectvolunteers, should establish what doses of boldine areneeded to reach plasma concentrations comparable tothose required for its antioxidant action. Upon establish-ing the latter, it would be necessary to also corroboratewhether the selected doses are indeed therapeuticallyeffective in preventing the development of the oxida-tive damage associated with various pathologies. Thesestudies should be complemented by further work pursu-ing the characterization of both the pharmacokinetics andantioxidant effects associated with the repeated admin-istration of boldine. The latter are also necessary in orderto be sure there are no side or non desirable effects ofboldine in humans (Tables 4 and 5).
References
[1] M.J. Magistretti, Remarks on the pharmacological examinationof plant extracts, Fitoterapia 51 (1980) 67–79.
[2] H. Speisky, B.K. Cassels, Boldo and boldine: an emergingcase of natural drug development, Pharmacol. Res. 29 (1994)1–12.
f thetract,
B.on,,
of alun-
L.and
.
iso-d
[ance15–
[ itro
[12] O.I. Aruoma, Free radicals, oxidative stress and antioxidants inhuman health and disease, J. Am. Oil. Chem. Soc. 75 (1998)199–212.
[13] K. Bauerova, A. Bezek, Role of reactive oxygen and nitrogenspecies in etiopathogenesis of rheumatoid arthritis, Gen. Physiol.Biophys. 18 (1999) 15–20.
[14] R. Kupfer, L.D. Dwyer-Nield, A.M. Malkinson, J.A. Thompson,Lung toxicity and tumour promotion by hydroxylated deriva-tives of 2,6-di-tert-butyl-4-methylphenol (BHT) and 2-tert-butyl-4-methyl-6-iso-propylphenol: correlation with quinone methidereactivity, Chem. Res. Toxicol. 15 (2002) 1106–1112.
[15] M. Saito, H. Sakagami, S. Fujisawa, Cytotoxicity and apopto-sis induction by butylated hydroxyanisole (BHA) and butylatedhydroxytoluene (BHT), Anticancer Res. 23 (2003) 4693–4701.
[16] G. Galati, P.J. O’Brien, Potential toxicity of flavonoids and otherdietary phenolics: significance for their chemopreventive and anti-cancer properties, Free Radic. Biol. Med. 37 (2004) 287–303.
[17] H. Speisky, B.K. Cassels, E.A. Lissi, L.A. Videla, Antioxidantproperties of the alkaloid boldine in systems undergoing lipidperoxidation and enzyme inactivation, Biochem. Pharmacol. 41(1991) 1575–1581.
[18] A. Valenzuela, S. Nieto, B.K. Cassels, H. Speisky, Inhibitoryeffect of boldine on fish oil oxidation, J. Am. Oil. Chem. Soc.68 (1991) 935–937.
[19] E.A. Lissi, T. Caceres, L.A. Videla, Visible chemiluminescencefrom rat brain homogenates undergoing autoxidation. Part I: effectof additives and products accumulation, J. Free Radic. Biol. Med.2 (1986) 63–69.
[20] J.L. Rıos, R. Cholbi, A.I. Huguet, A. Mora, S. Manez, M. Paya,M.J. Alcaraz, Effects of benzylisoquinolines on lipid peroxidationand superoxide scavenging, Planta Med. 56 (1990) 645–646.
[22] A. Ubeda, C. Montesinos, M. Paya, M.J. Alcaraz, Iron-reducingsome
993)
[ vercol.
[ tec-in
2000)
[ .S.rane
s in
[ liverme
[ hy,ug-
[ ela,en-07.
[ tiva-are
000)
[3] A. Ruegger, Neue alkaloide ausPeumus boldus Molina, Helv.Chim. Acta 48 (1959) 754–762.
[4] C. Van Hulle, P. Braeckman, R. Van Severen, Influence opreparation technique on the boldine content of boldo dry exJ. Pharm. Belg. 38 (1983) 97–100.
[5] M. Quezada, M. Asencio, J.M. del Valle, J.M. Aguilera,Gomez, Antioxidant activity of crude extract, alkaloid fractiand flavonoid fraction from boldo (Peumus boldus Molina) leavesJ. Food Sci. 69 (2004) 371–376.
[6] M. Gotteland, J. Espinoza, B.K. Cassels, H. Speisky, Effectdry boldo extract on oro-cecal intestinal transit in healthy voteers, Rev. Med. Chil. 123 (1995) 955–960.
[7] G. Schmeda-Hirschmann, J.A. Rodrıguez, C. Theoduloz, S.Astudillo, G.E. Feresin, A. Tapia, Free-radical scavengersantioxidants fromPeumus boldus Mol. (“Boldo”), Free RadicRes. 37 (2003) 447–452.
[8] A. Urzua, P. Acuna, Alkaloids from the bark ofPeumus boldus,Fitoterapia 54 (1983) 175–177.
[9] P. Pietta, P. Mauri, E. Manera, P. Ceva, Determination ofquinoline alkaloids fromPeumus boldus by high-pressure liquichromatography, J. Chromatogr. 457 (1988) 442–445.
10] H. Speisky, B.K. Cassels, S. Nieto, A. Valenzuela, L.J. Nunez-Vergara, Determination of boldine in plasma by high-performliquid chromatography, J. Chromatogr. 612 (1993) 3319.
11] I. Jimenez, H. Speisky, Biological disposition of boldine: in vand in vivo studies, Phytother. Res. 14 (2000) 254–260.
and free-radical-scavenging properties of apomorphine andrelated benzylisoquinolines, Free Radic. Biol. Med. 15 (1159–167.
23] A.I. Cederbaum, E. Kukielka, H. Speisky, Inhibition of rat limicrosomal lipid peroxidation by boldine, Biochem. Pharma44 (1992) 1765–1772.
25] Y.C. Youn, O.S. Kwon, E.S. Han, J.H. Song, Y.K. Shin, CLee, Protective effect of boldine on dopamine-induced membpermeability transition in brain mitochondria and viability losPC12 cells, Biochem. Pharmacol. 63 (2002) 495–505.
26] P. Kringstein, A.I. Cederbaum, Boldine prevents humanmicrosomal lipid peroxidation and inactivation of cytochroP4502E1, Free Radic. Biol. Med. 18 (1995) 559–563.
27] R. Kubinova, M. Machala, K. Minksova, J. Neca, V. SucChemoprotective activity of boldine: modulation of drmetabolizing enzymes, Pharmazie 56 (2001) 242–243.
28] B.K. Cassels, M. Asencio, P. Conget, H. Speisky, L.A. VidE.A. Lissi, Structure–antioxidative activity relationships in bzylisoquinoline alkaloids, Pharmacol. Res. 31 (1995) 103–1
29] I. Jimenez, E.A. Lissi, H. Speisky, Free-radical-induced inaction of lysozyme and carbonyl residue generation in proteinnot necessarily associated, Arch. Biochem. Biophys. 381 (2247–252.
16 P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17
[30] E. Mendez, J. Sanhueza, S. Nieto, H. Speisky, A. Valenzuela,Fatty acid composition, extraction, fractionation, and stabilizationof Bullfrog (Rana catesbeiana) oil, J. Am. Oil. Chem. Soc. 75(1998) 67–71.
[31] A. Ganga, S. Nieto, J. Sanhueza, C. Romo, H. Speisky, A. Valen-zuela, Concentration and stabilization ofn-3 poly-unsaturatedfatty acids from sardine oil, J. Am. Oil. Chem. Soc. 75 (1998)733–736.
[32] E. Silva, M. Jopia, A.M. Edwards, E. Lemp, J.R. De la Fuente,E. Lissi, Protective effect of boldo and tea infusions on the visi-ble light-mediated pro-oxidant effects of Vitamin B2, riboflavin,Photochem. Photobiol. 75 (2002) 585–590.
[33] A.L. Zanocco, E. Lemp, G. Gunther, A kinetic study of the reac-tion between boldine and singlet oxygen [O2(1�g)], J. Chem.Soc. Perkin Trans. 2 (1997) 1299–1302.
[34] H.R. Griffiths, P. Mistry, K.E. Herbert, J. Lunec, Molecular andcellular effects of ultraviolet light-induced genotoxicity, Crit. Rev.Clin. Lab. Sci. 35 (1998) 189–237.
[35] M.E. Hidalgo, I. Gonzalez, F. Toro, E. Fernandez, H. Speisky,I. Jimenez, Boldine as a sunscreen, its photoprotector capacityagainst UVB radiation, Cosmet. Toiletries 113 (1998) 59–66.
[36] F. Rancan, S. Rosan, K. Boehm, E. Fernandez, M.E. Hidalgo, W.Quihot, C. Rubio, F. Boehm, H. Piazena, U. Oltmanns, Protectionagainst UVB irradiation by natural filters extracted from lichens,J. Photochem. Photobiol. B 68 (2002) 133–139.
[37] F. Loghin, A. Chagraoui, M. Asencio, E. Comoy, H. Speisky,B.K. Cassels, P. Protais, Effects of some antioxidative aporphinederivatives on striatal dopaminergic transmission and on MPTP-induced striatal dopamine depletion in B6CBA mice, Eur. J.Pharm. Sci. 18 (2003) 133–140.
[38] L. Mili an, R. Estelles, B. Abarca, R. Ballesteros, M.J. Sanz, M.A.Blazquez, Reactive oxygen species (ROS) generation inhibitedby aporphine and phenanthrene alkaloids semi-synthesized fromnatural boldine, Chem. Pharm. Bull. 52 (2004) 696–699.
[48] M. Gotteland, I. Jimenez, O. Brunser, L. Guzman, S. Romero,B.K. Cassels, H. Speisky, Protective effect of boldine in experi-mental colitis, Planta Med. 63 (1997) 311–315.
[49] T. Ohkuwa, Y. Sato, M. Naoi, Hydroxyl radical formation indiabetic rats induced by streptozotocin, Life Sci. 56 (1995)1789–1798.
[50] N. Santanam, M. Penumetcha, H. Speisky, S. Parthasarathy, Anovel alkaloid antioxidant, boldine and synthetic antioxidant,reduced form of RU486, inhibit the oxidation of LDL in vitroand atherosclerosis in vivo in LDLR−/− mice, Atherosclerosis173 (2004) 203–210.
[51] C.M. Teng, C.M. Hsueh, Y.L. Chang, F.N. Ko, S.S. Lee, K.C. Liu,Antiplatelet effects of some aporphine and phenanthrene alkaloidsin rabbits and man, J. Pharm. Pharmacol. 49 (1997) 706–711.
[52] J. Hu, H. Speisky, I.A. Cotgreave, The inhibitory effects of bol-dine, glaucine, and probucol on TPA-induced down regulationof gap junction function: relationships to intracellular peroxides,protein kinase C translocation, and connexin 43 phosphorylation,Biochem. Phammacol. 50 (1995) 1635–1643.
[53] A. Morello, I. Lipchenca, B.K. Cassels, H. Speisky, J. Aldunate, Y.Repetto, Trypanocidal effect of boldine and related alkaloids uponseveral strains ofTrypanosoma cruzi, Comp. Biochem. Physiol.Pharmacol. Toxicol. Endocrinol. 107 (1994) 367–371.
[54] R.R. Moreira, I.Z. Carlos, W. Vilega, Release of intermediatereactive hydrogen peroxide by macrophage cells activated by nat-ural products, Biol. Pharm. Bull. 24 (2001) 201–204.
[55] R. Gonzalez-Cabello, H. Speisky, R. Bannach, A. Valenzuela, J.Feher, P. Gergely, Effects of boldine on cellular immune functionsin vitro, J. Investig. Allerg. Clin. Immunol. 4 (1994) 139–145.
[56] S. Philipov, N. Ivanovska, P. Nikolova, Glaucine analogues asinhibitors of mouse splenocyte activity, Pharmazie 53 (1998)694–698.
[57] J. Delourme, Action intestinale de la boldine, C. R. Hebd. SeancesAcad. Sci. 229 (1949) 953–955.
[58] E. Gleyses, Le boldo,Peumus boldus Molina: etude botanique,ues
[ e
[ a-
[.
[
[ h-rus,
[ ular, Jpn.
[ uralntral.
[ ra,
. 119
[ eal-
allylsecoboldine as a novel antioxidant against peroxidativeage, Eur. J. Pharmacol. 303 (1996) 129–139.
40] H. Jaeschke, Mechanisms of oxidant stress-induced acuteinjury, Proc. Soc. Exp. Biol. Med. 209 (1995) 104–111.
41] Y. Sato, S. Kamo, T. Takahashi, Y. Suzuki, Mechanismfree radical-induced hemolysis of human erythrocytes: heysis by water-soluble radical initiator, Biochemistry 34 (198940–8949.
42] I. Jimenez, A. Garrido, R. Bannach, M. Gotteland, H. SpeProtective effects of boldine against free radical-induced erycytes lysis, Phytother. Res. 14 (2000) 339–343.
43] K. Chan, D. Truong, N. Shangari, P.J. O’Brien, Drug indumitochondrial toxicity, Expert Opin. Drug Metab. Toxicol. 1(2005).
44] M.C. Lanhers, M. Joyeux, R. Soulimani, J. Fleurentin, M. SaF. Mortier, C. Younos, J.M. Pelt, Hepatoprotective and ainflammatory effects of a traditional medicinal plant of ChPeumus boldus, Planta Med. 57 (1991) 110–115.
45] R. Bannach, A. Valenzuela, B.K. Cassels, L.J. Nunez-VergaraH. Speisky, Cytoprotective and antioxidant effects of bolon tert-butyl-hydroperoxide-induced damage to isolated hepcytes, Cell Biol. Toxicol. 12 (1996) 89–100.
46] G. Plaa, Chlorinated methanes and liver injury: highlights opast 50 years, Annu. Rev. Pharmacol. Toxicol. 40 (2000) 42
47] N. Backhouse, C. Delporte, M. Givernau, B.K. Cassels, A. Vazuela, H. Speisky, Anti-inflammatory and antipyretic effectboldine, Agents Actions 42 (1994) 114–117.
chimique, pharmacologique, These de Sciences Pharmaceutiqet Biologiques, Universite Paris XI, Paris, 1986.
59] H. Speisky, J.A. Squella, L.J. Nunez-Vergara, Activity of boldinon rat ileum, Planta Med. 57 (1991) 519–522.
60] J.L. Delso-Jimeno, Colereticos y colagogos, un estudio farmcologico de la hoja de boldo, Anales. Inst. Farmacol. Espana 5(1956) 395–441.
61] M.C. Levy-Appert-Collin, J. Levy, Sur quelques preparationsgaleniques de feuilles de boldo (Peumus boldus, momiacees), JPharm. Belg. 32 (1977) 13–22.
62] H. Kreitmair, Pharmakologische wirkung des alkaloids ausPeu-mus boldus Molina, Pharmazie 7 (1952) 507–511.
63] M.D. Ivorra, F. Martınez, A. Serrano, P. D’Ocon, Different mecanism of relaxation induced by aporphine alkaloids in rat uteJ. Pharm. Pharmacol. 45 (1993) 439–443.
64] J.J. Kang, Y.W. Cheng, W.M. Fu, Studies on neuromuscblockade by boldine in the mouse phrenic nerve-diaphragmJ. Pharmacol. 76 (1998) 207–212.
65] B. Hue, H. Le Corronc, B. Kuballa, R. Anton, Effects of the natalkaloid boldine on cholinergic receptors of the insect cenervous system, Pharm. Pharmacol. Lett. 3 (1994) 169–172
66] S. Chulia, J. Moreau, E. Naline, M.A. Noguera, M.D. IvorM.P. D’Ocon, C. Advenier, The effect ofS-(+)-boldine on the�1-adrenoceptor of the guinea-pig aorta, Br. J. Pharmacol(1996) 1305–1312.
67] M.D. Ivorra, S. Chulia, C. Lugnier, M.P. D’Ocon, Selectivaction of two aporphines at�1-adrenoceptors and potenti
P. O’Brien et al. / Chemico-Biological Interactions 159 (2006) 1–17 17
operated Ca2+ channels, Eur. J. Pharmacol. 231 (1993) 165–174.
[68] S. Martınez, Y. Madrero, M. Elorriaga, M.A. Noguera, B.K.Cassels, E. Sobrazo, M.P. D’Ocon, M.D. Ivorra, Halogenatedderivatives of boldine with high selectivity for�1A-adrenoceptorsin rat cerebral cortex, Life Sci. 64 (1999) 1205–1214.
[69] M. Eltze, T. Grebe, M.C. Michel, P. Czyborra, B. Ullrich, Affinityprofile at�1- and�2-adrenoceptor subtypes and in vitro cardio-vascular actions of (+)-boldine, Eur. J. Pharmacol. 443 (2002)151–168.
[70] J.J. Kang, Y.W. Cheng, Effects of boldine on mouse diaphragmand sarcoplasmic reticulum vesicles isolated from skeletal mus-cle, Planta Med. 64 (1998) 18–21.
[71] Y. Madrero, M. Elorriaga, S. Martınez, M.A. Noguera, B.K. Cas-sels, M.P. D’Ocon, M.D. Ivorra, A possible structural determinantof selectivity of boldine and derivates for the�1A-adrenoceptorsubtype, Br. J. Pharmacol. 119 (1996) 1563–1568.
[72] G. Zetler, Neuroleptic-like, anticonvulsant and antinociceptiveeffects of aporphine alkaloids: bulbocapnine, corytuberine, bol-dine and glaucine, Arch. Int. Pharmacodyn. Ther. 296 (1988)255–281.
[73] M. Asencio, B. Delaquerriere, B.K. Cassels, H. Speisky, E.Comoy, P. Protais, Biochemical and behavioral effects of bol-
dine and glaucine on dopamine systems, Pharmacol. Biochem.Behav. 62 (1999) 7–13.
[74] E.M. Sobarzo-Sanchez, J. Arbaoui, P. Protais, B.K. Cassels, Halo-genated boldine derivatives with enhanced monoamine receptorselectivity, J. Nat. Prod. 63 (2000) 480–484.
[75] P.R. Moreno, V.M. Vargas, H.H. Andrade, A.T. Henriques, J.A.Henriques, Genotoxicity of the boldine aporphine alkaloid inprokaryotic and eukaryotic organisms, Mutat. Res. 260 (1991)145–152.
[76] D.C. Tavares, C.S. Takahashi, Evaluation of the genotoxic poten-tial of the alkaloid boldine in mammalian cell systems in vitroand in vivo, Mutat. Res. 321 (1994) 139–145.
[77] E.R. Almeida, A.M. Melo, H. Xavier, Toxicological evalua-tion of the hydro-alcohol extract of the dry leaves ofPeu-mus boldus and boldine in rats, Phytother. Res. 14 (2000) 99–102.
[78] S. Monzon, A. Lezaun, D. Saenz, Z. Marquinez, N. Bernedo, O.Uriel, C. Colas, F. Duce, Anaphylaxis to boldo infusion, a herbalremedy, Allergy 59 (2004) 1019–1020.
[79] F. Piscaglia, S. Leoni, A. Venturi, F. Graziella, G. Donati, L.Bolondi, Caution in the use of boldo in herbal laxatives: acase of hepatotoxicity, Scand J. Gastroenterol. 40 (2005) 236–239.
