Mechanism of Action of St John’s Wort in Depressionhomepages.gac.edu/~jwotton2/PSY385/mechanism in st. johns wort.pdf · Mechanism of Action of St John’s Wort 541 and, thus, it

CNS Drugs 2003; 17 (8): 539-562LEADING ARTICLE 1172-7047/03/0008-0539/$30.00/0

Adis Data Information BV 2003. All rights reserved.

Mechanism of Action of St John’sWort in DepressionWhat is Known?

Veronika Butterweck

Institute of Pharmacology and Toxicology, Universitatsklinikum Munster, Munster, Germany

Extracts of Hypericum perforatum L. (St John’s wort) are now successfullyAbstractcompeting for status as a standard antidepressant therapy. Because of this, greateffort has been devoted to identifying the active antidepressant compounds in theextract. From a phytochemical point of view, St John’s wort is one of thebest-investigated medicinal plants. A series of bioactive compounds has beendetected in the crude material, namely flavonol derivatives, biflavones, proantho-cyanidines, xanthones, phloroglucinols and naphthodianthrones.

Although St John’s wort has been subjected to extensive scientific studies inthe last decade, there are still many open questions about its pharmacology andmechanism of action. Initial biochemical studies reported that St John’s wort isonly a weak inhibitor of monoamine oxidase-A and -B activity but that it inhibitsthe synaptosomal uptake of serotonin, dopamine and noradrenaline (norepine-phrine) with approximately equal affinity. However, other in vitro binding assayscarried out using St John’s wort extract demonstrated significant affinity foradenosine, GABAA, GABAB and glutamate receptors. In vivo St John’s wortextract leads to a downregulation of β-adrenergic receptors and an upregulation ofserotonin 5-HT2 receptors in the rat frontal cortex and causes changes in neuro-transmitter concentrations in brain areas that are implicated in depression. Instudies using the rat forced swimming test, an animal model of depression, StJohn’s wort extracts induced a significant reduction of immobility. In otherexperimental models of depression, including acute and chronic forms of escapedeficit induced by stressors, St John’s wort extract was shown to protect rats fromthe consequences of unavoidable stress. Recent neuroendocrine studies suggestthat St John’s wort is involved in the regulation of genes that controlhypothalamic-pituitary-adrenal axis function. With regard to the antidepressanteffects of St John’s wort extract, many of the pharmacological activities appear tobe attributable to the naphthodianthrone hypericin, the phloroglucinol derivativehyperforin and several flavonoids.

This review integrates new findings of possible mechanisms that may underliethe antidepressant action of St John’s wort and its active constituents with a largebody of existing literature.

540 Butterweck

1. St John’s Wort as an Antidepressant Since the meta-analysis appeared,[12] severalmore double-blind, randomised, controlled trials

Mental illness imposes a tremendous burden onhave been published that have indicated the efficacy

the Western world. Mental disorders can strike earlyof St John’s wort in depression.[9,11,13-15] However, ain life, and they are increasing in incidence in anrecent large-scale negative study by Shelton et al.[16]

aging population experiencing neurodegenerativeraised concerns about the effectiveness of the herb.diseases. The search for new and more effectiveIt should be emphasised that in this clinical trial,therapeutic agents includes the study of plants usedpatients with severe depression were included, anin traditional medicine systems to treat mental disor-indication for which St John’s wort is not approvedders. In the last 10–15 years, phytomedicines basedin any country (despite an initial trial that suggestedon extracts from the herb of Hypericum perforatumefficacy in this population[14]). A large-scale studyL. (St John’s wort, Clusiaceae) have gained wide-coordinated by researchers of the National Institutesspread popularity as “the Prozac of herbs”.[1]

of Mental Health (NIMH) in the US tried to addressOver the past 2000 years, St John’s wort has beensome of the unanswered questions.[17] St John’s wortsingled out for its diverse medicinal properties bywas compared with placebo and an SSRI (sertraline)eminent medical writers and folk healers; the prima-

ry ancient medical herbalists, including Hip- in 340 patients for 8 weeks initially and a further 4pocrates, Pliny, Dioscurides and Galen, wrote about months for those who responded positively. Thethe medicinal properties of St John’s wort, noting its study failed to support the efficacy of St John’s wortuse as a wound-healing, diuretic and antimalarial in severe depression. While the authors found noagent. It was also used in a number of European evidence for a superior effect of St John’s wortcountries for the treatment of neuralgia, menopausal relative to placebo, sertraline could also not be dif-neurosis, anxiety and depression and as a nerve tonic ferentiated from placebo on the primary efficacy(for review see Bombardelli and Morazzoni[2]). To- measures, which casts doubt on the validity of theday, alcoholic extracts produced from the upper study. Whether the doses used in this trial werethird of the flowering plant are used for the treat- sufficiently high for treating severe depression isment of mild to moderate depression. Preparations questionable, since the response rates from bothof St John’s wort are some of the most prescribed

groups were relatively low. Thus, the results of themedicines in Germany; the number of prescriptions

trials[16,17] involving severely depressed patients arefor them have approximately tripled since 1993 (in

unconvincing, and further studies are required to1999 approximately 130 million daily doses wereinvestigate the effectiveness of St John’s wort inprescribed).[3,4] In the US, St John’s wort is increas-comparison with conventional antidepressants iningly used as an over-the-counter remedy for thepatients with severe depression.treatment of depression.[5]

Comparisons of different St John’s wort prepara-1.1 Evidence of Efficacy tions and dose-finding studies are still lacking. The

trial by Laakmann et al.[15] is the only investigationThe antidepressant efficacy of St John’s wortcomparing two different extracts of St John’s wort.extracts has been confirmed in several clinical stud-It showed that an extract with 5% hyperforin wasies.[6-11] A meta-analysis that was published in 1996significantly more effective, in terms of reducing theby Linde et al.[12] of trials of St John’s wort conclud-Hamilton Depression Rating Scale score, than aned that there is good evidence that St John’s wort isextract with 0.5% hyperforin. This finding suggest-superior to placebo in treating mild to moderateed that hyperforin may be of relevance for the clin-depression. However, the authors emphasised theical efficacy of the extract. However, data on theneed for more studies with larger samples andother constituents (such as flavonoids, hypericinsparticularly for trials comparing St John’s wort withand biflavones) in the extracts were not presentedconventional antidepressants.

Adis Data Information BV 2003. All rights reserved. CNS Drugs 2003; 17 (8)

Mechanism of Action of St John’s Wort 541

and, thus, it cannot be excluded that constituents 3. St John’s Wort and the CommonHypotheses of Antidepressant Actionother than hyperforin may have clinical efficacy.[18]

It can be concluded that several clinical trialsstrengthen the case for St John’s wort as an antide- 3.1 Inhibition of Monoamine Oxidasepressant, but more work needs to be done to answer

Several early in vitro experiments with St John’sthe remaining questions. In particular, the identifica-wort focused on pathways that alter monoaminetion of the effects of different (analytically wellneurotransmission in the CNS. Initial reports sug-

characterised) preparations, administration sched-gested that inhibition of monoamine oxidase (MAO)

ules, optimal doses and treatment durations would– the enzyme that is responsible for the catabolism

be valuable.of biogenic amines – is the main mechanism ofantidepressant action of St John’s wort extract.

2. Constituents Based on the work of Suzuki et al.,[31] hypericin wasconsidered to be an inhibitor of both MAO type Aand type B. Demisch et al.[32] and Sparenberg etFrom a phytochemical point of view, St John’sal.[33] confirmed the inhibitory effect on MAO of thewort is one of the best-investigated medicinal plants.St John’s wort extract but not of hypericin. TheA series of bioactive compounds has been detectedauthors found MAO inhibitory effects by the flavo-in the crude material, namely phenylpropanes, fla-noids and xanthones.vonol derivatives, biflavones, proanthocyanidines,

Bladt and Wagner[34] investigated the effect onxanthones, phloroglucinols, some amino acids,MAO of six fractions from St John’s wort extract.naphthodianthrones and essential oil constituentsInhibition of MAO-A could be shown with the total(for reviews see Bombardelli and Morazzoni,[2]

extract at a relatively high concentration of 10–3Nahrstedt and Butterweck[19] and Nahrstedt[18])mol/L (based on a molecular weight [MW] of quer-[structures are shown in figure 1 and figure 2].citrin of 449). In 1994, Thiede and Walper[35] inves-

Although St John’s wort has been subjected to tigated the effects of St John’s wort extract, hyper-extensive scientific studies in the last decade, there icin and some St John’s wort fractions on the activi-are still many open questions about its pharmacolo- ty of MAO and catechol-O-methyltransferasegy and mechanism of action. In fact, the active (COMT). Inhibition of MAO was observed with Stconstituents are not fully known. Recently, the anti- John’s wort extract 10–4 mol/L (based on an averagedepressant activity of St John’s wort extracts has MW of 500) and hypericin 10–3 mol/L. Only the Stbeen variously attributed to the phloroglucinol de- John’s wort extract inhibited COMT. The isolatedrivative hyperforin,[22-25] to the naphthodianthrones fractions were less active in both enzyme assays.hypericin and pseudohypericin[20,26-28] and to several The authors concluded that the clinically provenflavonoids.[29,30] The role and mechanisms of these antidepressant activity of St John’s wort cannot bedifferent compounds are still a matter of debate. explained in terms of MAO inhibition. These find-However, based on recent results, it seems that the ings were later supported by Cott[36] and Muller andprevailing simplistic view of one plant → one active coworkers.[37] Both authors confirmed the rathercompound → one mechanism of action is incorrect. weak potency of St John’s wort extract as an inhibi-It is more likely that the multiple bioactive com- tor of MAO-A and -B.pounds contribute to the antidepressant activity of Suzuki et al.[31] were the first to find thatthe crude plant extract in a complex manner. micromolar concentrations of hypericin could irre-

This review focuses on the present knowledge versibly inhibit MAO-A and -B activity in vitro.about the mechanism of action of St John’s wort and However, the progress made in preparation and ana-its active constituents in depression, with regard to lysis techniques since 1984 has shown that thethe most recent literature. hypericin used in these experiments was impure and


542B

utterweck

A

dis D

ata

Info

rma

tion

BV 2003. A

ll righ

ts rese

rved

.C

NS D

rug

s 2003; 17 (8)

Group of bioactive compound ConstituentApproximate % amount in a crude St John's wort extract*

Flavonoids 1. 0.2–0.62. 0.2–0.43. 0.5–0.84. 2–45. 2–36. 0.4–0.05

Phloroglucinols 0.9–5

Naphthodianthrones

1. R = H2. R = α-L-rhamnosyl3. R = β-D-glucosyl4. R = β-D-galactosyl5. R = β-D-rutinosyl6. R = β-D-glucuronide

QuercetinQuercitrineIsoquercitrinHyperosideRutinMiquelianin

Hyperforin

1. R = CH3

2. R = CH2OHHypericinPseudohypericin

1. 0.08–0.202. 0.15–0.40

Structure

O

OH

OH

OR

OHO

HO

O OH

OO

CH3HO

HO

O

O

OH

OH

OH

R

OH

Fig. 1. Chemical structures of the major constituents of St John’s wort: flavonoids, naphthodianthrones and phloroglucinols. * Data on content of bioactive compounds in crudeextract from Butterweck et al.,[20] Nahrstedt and Butterweck[19] and Jurgenliemk and Nahrstedt.[21]

Mechanism

of Action of St John’s W

ort543

A

dis D

ata

Info

rma

tion

BV 2003. A

ll righ

ts rese

rved

.C

NS D

rug

s 2003; 17 (8)

Biflavones 1. 0.2–0.62. 0.01–0.05

Phenylpropanes <1

Xanthones <0.0004

Proanthocyanidines Total amount ofproanthocyanidinesincluding B2 ~8.8

1. I3, II8-biapigenin2. I3′, II8-biapigenin = amentoflavone

Chlorogenic acid

1,3,6,7-Tetrahydroxyxanthone

Procyanidin B2

OHO

OH O

O HO

O

HO OH

OH

3

3′

8

OH

OH

O

O

OH

OH

HOOC 1

5

3

O

O

HO

HO

OH

OH

1

36

7

O

OH

OH

OH

OH

HO

O

OH

OH

OH

OH

HO

Group of bioactive compound ConstituentApproximate % amount in a crude St John's wort extract*

Structure

Fig. 2. Chemical structures of the major constituents of St John’s wort: biflavones, phenylpropanes, xanthones and proanthocyanidins. * Data on content of bioactive compounds incrude extract from Butterweck et al.,[20] Nahrstedt and Butterweck[19] and Jurgenliemk and Nahrstedt.[21]

544 Butterweck

contained at least 20% of other constituents of the 3.2 Inhibition of Synaptosomal Reuptakeof Aminesextracts; among these, the flavonoids are the most

noticeable. In fact, the MAO inhibitory effects ofReuptake into nerve terminals is the main mecha-hypericin could not be confirmed in subsequent

nism for terminating the actions of synaptically re-studies.[32,34-36] It was suggested that the flavonoidleased biogenic amines such as serotonin, norad-derivatives, therefore, might be responsible for therenaline (norepinephrine) and dopamine.[40-42] Inhi-

inhibitory effects on MAO, as the chemical structure bition of the reuptake, via blockade of distinctof those compounds is similar to that of synthetic transporters, is a common action of a variety ofMAO inhibitors.[38] Indeed, the flavonoids querce- antidepressants, including tricyclic antidepressantstin, luteolin and kaempferol have been shown to (TCAs) and SSRIs,[43,44] and results in an increase of

monoamine activity at synapses.[45,46]inhibit MAO activity in vitro.[35] However, the con-Research with St John’s wort also focused oncentration in St John’s wort of these substances,

inhibition of synaptosomal reuptake of serotoninespecially that of the aglycones, is too low to beand/or noradrenaline; it was assumed that this mightresponsible for the therapeutic efficacy of St John’spoint to an initial mechanism of action of the plant

wort extracts.[18,21] Chatterjee et al.[22] investigatedextract similar to that of most other antidepressant

the phloroglucinol derivative hyperforin and found drugs. All data regarding monoamine reuptake arethat it did not inhibit MAO-A or -B activities. summarised in table II.

The first studies in this field were performed byData from all the above-mentioned studies arePerovic and Muller.[50] The authors showed that Stsummarised in table I. In conclusion, although sev-John’s wort extract is a rather potent inhibitor oferal studies demonstrate that MAO inhibitors aresynaptosomal serotonin reuptake. This finding waspresent in St John’s wort, their relatively high inhib-confirmed by Muller and Schafer,[39] who found that

itory concentrations do not suggest any relevance of St John’s wort extract inhibited synaptosomal reup-MAO-A or -B inhibition to the antidepressant effect take of serotonin, noradrenaline and dopamine withof St John’s wort. These results suggest that other quite similar potencies.mechanisms of action contribute to the antidepres- In order to identify the major reuptake-inhibitingsant activity of the herb. compound of St John’s wort, initial research focused

Table I. Monoamine oxidase (MAO) inhibition by St John’s wort extract and various compounds

References MAO-A and -B inhibition (µmol/L range)

Hydroalcoholic St John’s wort extract

Bladt & Wagner,[34] Chatterjee et al.,[22] Cott,[36] Demisch et al.,[32] Muller and Schafer,[39] +Muller et al.,[37] Sparenberg et al.,[33] Suzuki et al.,[31] Thiede & Walper[35]

Hypericin

Suzuki et al.[31] +

Bladt & Wagner,[34] Cott,[36] Demisch et al.,[32] Sparenberg et al.[33] –

Hyperforin

Chatterjee et al.[22] –

Quercitrin

Bladt & Wagner,[34] Demisch et al.,[32] Sparenberg et al.,[33] Thiede & Walper[35] +

Quercetin

Demisch et al.,[32] Sparenberg et al.,[33] Thiede & Walper[35] +

1,3,6,7-Tetrahydroxyxanthone

Demisch et al.,[32] Sparenberg et al.,[33] Suzuki et al.[31] ++ indicates inhibition; – indicates no effect.



on the lipophilic phloroglucinol derivativehyperforin. Later systematic studies included abroad spectrum of compounds. The first studies withhyperforin were performed by Chatterjee et al.[22]

and Muller and colleagues.[48] The authors foundthat hyperforin was capable of inhibiting the reup-take of all three monoamines with a potencycomparable to that of conventional serotonin andnoradrenaline reuptake inhibitors. Moreover, the au-thors noted that the ability of hyperforin to inhibitthe reuptake of serotonin, noradrenaline and dop-amine at a nanomolar concentration sets it apartfrom any known synthetic antidepressant. Subse-quent papers confirmed that hyperforin is a mono-amine reuptake inhibitor, but only at micromolarconcentrations.[47,49,52] In these studies, hypericinwas inactive in all monoamine systems. This findingwas also confirmed by Kientsch et al.[51] and Raf-fa.[53]

In a recent study, Wonnemann et al.[25] systemati-cally investigated the synaptosomal reuptake inhibi-tory properties of several constituents of St John’swort. The authors again identified hyperforin as themajor reuptake-inhibiting principle; however, it hasto be mentioned that this result was not obtainedusing the pure compound but rather by correlationanalysis of IC50 values (concentrations that produce50% inhibition of reuptake) of ten different StJohn’s wort extracts based on different amounts of

Table II. Comparison of the effects of different St John’s wort ex-tracts, pure compounds and standard synthetic antidepressants onreuptake of monoaminesa

Substance Serotonin Noradrenaline Dopaminereuptake (norepinephrine) reuptake

reuptake

St John’s wort extracts

Hydroalcoholic St +[22,37,39,47-50] +[22,37,39,48,49,51] +[22,37,39,48]

John’s wort extract

CO2 extract +[48,52] +[47,48] +[47,48]

Hyperforinb +[22,25,47-49,52]; +[22,25,48,49,52]; +[22,25,48,49,52]

–[51] –[51]

Adhyperforin +[25,52] +[25,52] +[25,52]

Hypericin –[50,51,53] –[48,51,53] –[48,51,53]

Hyperoside –[25] –[25] –[25]

Biapigenin –[25] –[25] –[25]

Quercitrin –[25] –[25] –[25]

Isoquercitrin –[25] –[25] –[25]

Rutin –[25] –[25] –[25]

Amentoflavone –[25] –[25] –[25]

Oligomeric +[25] +[25] +[25]

procyanidins

Standard antidepressants

Imipramine +[37,52] +[37,52] –[37,52]

Nomifensine –[37,52] +[37,52] +[37,52]

Fluoxetine +[52] +[52] –[52]

a This table presents an overview of investigations into effectson monoamine uptake, but it does not consider quantitativedifferences because the original data from different authorshad a different ‘cut-off’ for activity. Activities were observed ina nmol/L to µmol/L range.

b In the study by Wonnemann et al.,[25] various extracts withdifferent amounts of hyperforin were tested; pure hyperforinwas not tested.

+ indicates inhibition of uptake; – indicates no inhibition of uptake.hyperforin. Further, analytical data of additionalconstituents in the extracts, such as flavonoids or

transport proteins by procyanidins is not excludedbiflavones, are missing. Thus, it cannot be excludedthat constituents other than hyperforin may have by the results of this study.caused the synaptosomal reuptake inhibition. Recently, Kientsch et al.[51] investigated the acute

Interestingly, in the same study, the authors[25] actions of the St John’s wort extract ZE 117 onfound that hyperforin-free preparations were also neurotransmitter reuptake into rat brain cortexable to inhibit the reuptake systems in a weak to slices. The ZE 117 extract (standardised to 0.2%moderate manner. A fraction containing mostly hypericin, <0.1% hyperforin) was more selective foroligomeric procyanidins showed a weak to moder- the reuptake of noradrenaline than for serotonin.ate inhibition of synaptosomal reuptake of seroto-

Interestingly, none of these effects could benin, noradrenaline, dopamine, GABA and L-gluta-achieved with either hypericin or hyperforin in amate. Although these data are interesting, the impor-relevant dose range. The authors concluded that “thetant information of specific activity is missing:extract ZE 117 contains active, but as yet unknownprocyanidins are tannins and as such they precipitate

proteins.[54] However, the nonspecific inhibition of antidepressant principles with effects comparable to


546 Butterweck

those of TCAs”. Systematic studies confirming this interaction with the [3H]WIN 35,428 (a cocainehypothesis are required. analogue) binding site on the dopamine transporter

as seen for imipramine, nomifensine and fluoxetine.Buchholzer et al.[55] presented the first data onTherefore, the authors suggest a noncompetitiveinteractions of hyperforin with the cholinergic sys-interaction of hyperforin and adhyperforin with thetem. The authors found that hyperforin inhibiteddopamine transporter.high-affinity choline reuptake in rat synaptosomes.

These results are of special interest and could form Another interesting hypothesis regarding thethe basis for the investigation of the use of mechanism of synaptosomal reuptake is presentedhyperforin in cognitive disorders. by Muller et al.[24] and Singer et al.[23] The apparent

inhibition of serotonin reuptake observed with St3.3 Effects on Monoamine Transporters John’s wort extract and hyperforin in vitro might be

due to an increase in free intracellular sodium con-In order to explain the mechanism of inhibition of

centrations.[23] Additional data suggest interactionssynaptosomal reuptake, several researchers have fo-

of hyperforin with Na+ channels or Na+/H+ ex-cused on the effects of St John’s wort on monoamine

changers.[56] Such nonselective effects might ex-transporters in intact neural cells. The neurotrans-

plain why St John’s wort extracts and hyperforinmitter transporters have been proven to be important

blocked the synaptosomal reuptake of multipletargets for CNS drug discovery, particularly for

neurotransmitters.[22,23,47,56]

antidepressants.[40]

Recent work indicated that hyperforin may alsoGobbi et al.[47] reported that inhibition of seroto-influence calmodulin-dependent mechanisms.[57]

nin reuptake by a methanolic St John’s wort extractHyperforin was also reported to modulate severaland hyperforin is not due to a direct interaction with,ionic conductances in cerebellar Purkinje cells in-and blockade of, the serotonin transporter becausecluding a P-type calcium channel that is known to beboth compounds had no or only a very slight inhibi-involved in neurotransmitter release.[57] Krishtal ettory effect on [3H]citalopram binding. The authorsal.[58] also screened an ethanolic St John’s wortspeculated that the inhibitory effects onextract (hyperforin content approximately 5%) onsynaptosomal serotonin reuptake might be due to avarious voltage- and ligand-gated conductances.reserpine-like mechanism, because similar resultsThe plant extract inhibited almost all the ligand-were obtained with Ro-04-1284, a reserpine-likegated ion channels; quercitrin was detected as acompound, that also did not inhibit [3H]citaloprampotent inhibitor of adenosine triphosphate (ATP)-binding. Their hypothesis was confirmed by theinduced conductance; biapigenin inhibited both thefinding that St John’s wort extract and hyperforinacetylcholine- and ATP-induced conductance,induced a marked release of serotonin fromwhereas hyperoside blocked currents activated bysynaptosomes preloaded with [3H]serotonin. TheATP and α-amino-3-hydroxy-5-methyl-isox-authors concluded that the apparent inhibition ofazole-4-propionate (AMPA). Hypericin was inac-reuptake is an artefact due to the interaction of thetive in all cases. The authors concluded thathigh concentration of the tested compounds withhyperforin is not the only active substance of Stmonoamine storage vesicles.John’s wort and that the extract contains severalJensen et al.[52] found that the potency ofpotentially neuroactive molecules, such ashyperforin and adhyperforin in terms of inhibition ofbiapigenin, hyperoside and quercitrin.monoamine reuptake was comparable to that seen

The concentration of hyperforin (micromolarwith imipramine, nomifensine and fluoxetine. But inrange) used in these studies[57-59] was far higher thancontrast to the antidepressant drugs, the phlorogluci-plasma concentrations reached in vivo. In humannols potently inhibited all three transporter systems.volunteers receiving daily doses of St John’s wortInterestingly, the inhibition of dopamine uptake byextract 900mg (three 300mg doses), steady-statehyperforin and adhyperforin was not due to a direct



plasma concentrations of approximately 180 nmol/L (250–500 mg/kg orally). In this study, serotoninhave been measured.[60] Reported IC50 values for levels remained unchanged, whereas levels of thehyperforin as an inhibitor of synaptosomal reuptake serotonin metabolite 5-hydroxyindolacetic acidof serotonin have ranged from 120 to 3300 nmol/L. (5-HIAA) were significantly increased after treat-According to the results of the studies mentioned ment with 500 mg/kg of the plant extract; no effectsabove,[23,24,47,52] the blood concentrations of on noradrenaline levels were observed, and DOPAChyperforin in human volunteers after a daily dose of and HVA levels were significantly increased.St John’s wort extract 900mg are within a concen- In the study by Fornal et al.,[63] two commercialtration range needed to inhibit serotonin reuptake in preparations of St John’s wort (Jarsin3001,vitro, but the free drug concentration available for Hyperforat) failed to affect neuronal activity in theaction at central ion channels in vivo is unclear. dorsal raphe nucleus of awake cats after single-dose

Taken together, the in vitro effects of St John’s treatment, whereas fluoxetine and sertraline mark-wort and hyperforin clearly show that both are po- edly decreased neuronal activity by increasing thetent but nonspecific inhibitors of synaptosomal sero- synaptic availability of serotonin at inhibitory so-tonin, noradrenaline and dopamine, but that the pre- matodendritic 5-HT1A autoreceptors.cise mechanism relevant to therapeutic actions is Di Matteo et al.[64] used in vivo microdialysis tostill unknown. It is likely that other mechanisms investigate the effects of a CO2 extract (not furthermay also contribute to the overall antidepressant characterised) on dopamine, DOPAC, serotonin andaction of St John’s wort. Because all these data were 5-HIAA levels. At a dose of 1 mg/kg orally, theobtained in in vitro experiments, it is of interest to extract caused a slight but significant increase ofknow if these interactions with monoamine systems dopamine outflow both in the nucleus accumbensactually occur in vivo. and the striatum. No significant changes in serotonin

or 5-HIAA efflux were observed. Similar results3.4 Effects on Brain Monoamine Levels were reported by Rommelspacher and col-

leagues:[65] after single-dose administration of a3.4.1 Single-Dose Treatment methanolic St John’s wort extract (31.25–125 mg/kg

orally) or a CO2 extract (2.42–9.68 mg/kg orally),Calapai et al.[30] demonstrated that a methanolicdopamine and serotonin levels were increased in theextract of St John’s wort (125–500 mg/kg orally),nucleus accumbens.like fluoxetine, increased serotonin levels in the

cortex and noradrenaline and dopamine levels in the Franklin et al.[66] measured monoamine levels indiencephalon after a single dose in rats. Similar whole brain tissues after a single intraperitonealfindings were reported by Yu,[61] who observed an dose of a methanolic St John’s wort extract (50–100increase of serotonin in mouse hippocampal and mg/kg), hypericin (0.4 mg/kg) or hyperforin (9.3hypothalamic tissues after a single dose of a St mg/kg). While the extract (100 mg/kg) increasedJohn’s wort extract (extract not further character- levels of dopamine and serotonin, hypericin andised; 10–100 mg/kg orally). However, this author hyperforin had no effect on monoamine levels.noted that there were no significant changes of nor- Using the microdialysis technique, Kaehler etadrenaline, dopamine, 3,4-dihydroxyphenylacetic al.[67] and Philippu[68] found enhanced extracellularacid (DOPAC) or homovanillic acid (HVA) levels. levels of dopamine, noradrenaline, serotonin andThe author also reported unchanged caudate and glutamate in the rat locus coeruleus after a singlecortex serotonin levels. intraperitoneal injection of hyperforin 10 mg/kg.

Serdarevic et al.[62] determined neurotransmitter Levels of 5-HIAA, GABA, taurine, aspartate, serinelevels in whole mouse brains after single-dose treat- and arginine were not influenced. In the same study,ment with a methanolic St John’s wort extract a hyperforin-free St John’s wort extract also elevat-

1 The use of tradenames is for product identification purposes only and does not imply endorsement.


548 Butterweck

ed extracellular levels of catecholamines. In contrast weeks) administration, similar to imipramine (15to hyperforin, the extracellular level of serotonin mg/kg orally). All three drugs (St John’s wort ex-was decreased. Therefore, the author[68] concluded tract, hypericin and imipramine) increased serotoninthat the differing profiles of hyperforin and levels in the hypothalamus after long-term (8 weeks)hyperforin-free St John’s wort extract on extracellu- but not short-term (2 weeks) administration. Inter-lar transmitter levels point to the presence of an estingly, serotonin levels remained unchanged inadditional biologically active compound in St hippocampal tissues after both treatment paradigms.John’s wort. The level of noradrenaline was not changed after

Based on the results in the above-mentioned short-term (2 weeks) treatment with either imip-studies, the effects of single doses of St John’s wort ramine or St John’s wort extract, but it was reducedextract on monoamine levels seem to be relatively in the hippocampus after 2 weeks of daily adminis-inconsistent. These inconsistencies could be ex- tration of hypericin. This effect was still obviousplained by: (i) anatomically different brain regions after 8 weeks.assessed; (ii) different techniques (microdialysis,

Similar results after 2 weeks of daily treatmentbrain tissues, electrophysiological recordings); (iii)with a methanolic St John’s wort extract were re-differences in the phytochemical composition of theported by Franklin and Cowen:[72] the extract (20extracts (variation of plant material, methanolic orand 100 mg/kg intraperitoneally) had no effect onCO2 extract); (iv) different administration routesserotonin, dopamine and noradrenaline levels in rat(orally, intraperitoneally, intravenously); and (v)whole brain tissues. It can be speculated that adifferent extract doses.treatment period >2 weeks is required to detect

3.4.2 Longer Term Treatment significant changes in brain monoamines, becauseDespite the importance of data from single-dose independent of the administration route (Franklin

treatment studies, the therapeutic mode of action of and Cowen,[72] intraperitoneally vs Butterweck etSt John’s wort needs to be established in studies al.,[71] orally), the same methanolic extract caused nobased on repeated treatment. Data after repeated effects. Increasing the treatment period to 8 weeks,treatment are of special interest because of the class- Butterweck et al.[71] found that St John’s wort ex-ically observed time lag between onset of antide- tract and hypericin induced a significant reductionpressant drug therapy and reversal of the depressed of dopamine turnover in the hypothalamus at the endmood. Like imipramine, the onset of efficacy of St of the treatment period.John’s wort extract occurs typically after 2–3 weeks

The in vivo influence of St John’s wort on theof treatment, so it is believed that the medicationdopaminergic system has been shown by Winterhoffcauses delayed CNS adaptations.[69,70]

et al.[73] and Butterweck et al.[74] In in vitro experi-In a recent study, Rommelspacher et al.[65] show-

ments, Muller et al.[48] reported that a CO2 extracted that after 14 days of daily administration, aand a methanolic St John’s wort extract inhibitedmethanolic St John’s wort extract (31.25–125 mg/kgdopamine reuptake. Moreover, the authors foundorally) had no effects on dopamine levels in the ratthat hyperforin inhibited dopamine reuptake, where-nucleus accumbens, whereas a CO2 extractas the naphthodianthrone hypericin failed to inhibit(2.42–9.68 mg/kg orally) caused a clear increase.dopamine reuptake.Interestingly, after the same treatment period, both

A dopaminergic mode of action has also beenextracts significantly increased serotonin levels insuggested by Franklin et al.[66,75] and Franklin andthis brain region.Cowen.[72] The authors found that St John’s wortButterweck et al.[71] recently demonstrated that aextract increased plasma growth hormone levels andSt John’s wort extract and the naphthodianthronedecreased plasma prolactin levels in human volun-hypericin affected levels of monoamine neurotrans-

mitters and their metabolites after long-term (8 teers after acute administration. Dopaminergic path-



ways facilitate growth hormone release and sup- of action of antidepressant drugs.[82] The 5-HT1Apress prolactin secretion.[76] and 5-HT2 receptors are candidate targets for antide-

In conclusion, based on the neurochemical stud- pressants within the serotonergic system.ies, changes in adrenergic, dopaminergic and/or Teufel-Mayer and Gleitz[83] showed that long-serotonergic neurotransmission may play an impor- term administration (26 weeks) of a St John’s worttant role in mediating the antidepressant effects of extract (extract not further characterised) upregulat-long-term administration of St John’s wort. How- ed the number of postsynaptic 5-HT1A and 5-HT2Aever, further pharmacological in vivo studies of the receptors in rat brain with no changes in receptoreffects of long-term administration of St John’s wort

affinity. The dose of St John’s wort extract was 2700are necessary to support this hypothesis.

mg/kg. The relevance of these pharmacological ef-fects is therefore questionable, especially because

3.5 Effects on Monoamine Receptorsdoses above 1500 mg/kg have been proven to betoxic in rats.[84]

3.5.1 Adrenergic ReceptorsAn upregulation of 5-HT2 receptors in the ratA consequence of antidepressant treatment and

frontal cortex after 14 days of daily treatment withelevated levels of serotonin or noradrenaline is the240 mg/kg of a methanolic St John’s wort extractoccupation of monoamine receptors. One possibilitywas also reported by Muller et al.,[37] whereas a CO2is that persistent activation of these receptors leadsextract slightly decreased 5-HT2 receptor density.[48]

to slowly developing receptor adaptations that ex-Most conventional antidepressants are known toplain the delayed onset of therapeutic action ofdownregulate the expression of 5-HT2 receptors inantidepressant treatment.[77-79] Early studies demon-rat brain,[85,86] whereas electroconvulsive shockstrated that long- but not short-term administrationtreatment (ECT) causes a significant elevation ofof several types of antidepressants decreases β-ad-5-HT2 receptor density.[87] With respect to therenergic receptor ligand binding in relevant areasupregulation of 5-HT2 receptors seen by Muller etsuch as the cerebral cortex and hippocampus.[80,81]

al.[37] after treatment with St John’s wort extract, theThe question of whether β-adrenoceptor block-action of ECT on 5-HT2 receptors resembles that ofade contributes to the effect of St John’s wort wasthe plant extract. This conformity may indicate thatfirst addressed by the study of Muller et al.[37] TheSt John’s wort and ECT could act by a similar modeauthors showed that subacute treatment (14 days)

with a methanolic extract of St John’s wort (240 mg/ of action, although ECT is known to be effectivekg orally) significantly downregulated the number against severe depression whereas St John’s wort(Bmax) of rat cortical β-adrenoceptors without a extracts are indicated for mild to moderate depres-change in receptor affinity (KD). The same result – a sive disorders.downregulation of β-adrenoceptors (of 15%) – was In a recent study, Butterweck et al.[27] found thatachieved using a lipophilic extract obtained with long-term treatment (8 weeks) with imipramine (15supercritical CO2 (hyperforin content 38.8%).[48]

mg/kg orally), St John’s wort extract (500 mg/kgThe authors concluded that hyperforin must be con- orally) and hypericin (0.2 mg/kg orally) decreasedsidered a major active antidepressant component of

5-HT1A receptor messenger RNA (mRNA) expres-St John’s wort. However, the effects of other con-

sion in the CA1 of the hippocampus. In the samestituents of St John’s wort on β-adrenoceptor

study, the authors found that long- but not short-downregulation have not been investigated system-

term immobilisation stress downregulated 5-HT1Aatically.gene expression in the hippocampus. The significantdecrease in 5-HT1A gene expression is consistent3.5.2 Serotonin Receptorsacross various stress experiments.[88,89] ImipramineSerotonin has long been implicated in the biolog-and St John’s wort extract were unable to preventical basis of depression as well as in the mechanism


550 Butterweck

the stress-induced decrease in 5-HT1A receptor den- 3.6 Receptor Binding Profile

sity. Several hypotheses have been advanced to ex-Data suggest that the common action of antide- plain the therapeutic actions of St John’s wort ex-

tract on a molecular basis. In vitro as well as in vivopressants could be related to an effect on the regula-studies have implicated activity at serotonergic, nor-tion of the expression of serotonergic receptor sub-adrenergic, dopaminergic and opioid receptors intypes,[90,91] but there is considerable controversy inthe hypothesised mechanism of antidepressant ac-

the literature, particularly using receptor bindingtion. In this section, the in vitro data only will be

studies and behavioural experiments, with regard to summarised and discussed. The results from studiesthe effects of TCAs and SSRIs on 5-HT1A and of receptor binding are summarised in table IV.5-HT2A receptors, with upregulation, downregula- Most in vitro research has focused on the crude St

John’s wort extract.[37,39,47,96,98] There is a paucity oftion and no effects being reported.[92-95] None ofin vitro data on single substances, and most authorsthese mechanisms is completely satisfactory as afocused on either a few selected sub-common antidepressant mechanism of action. Thestances[22-24,53,56,100] or individual receptor

fact that long-term use of St John’s wort produced systems.[47,96,97] One recent study examined the ma-changes in 5-HT1 and 5-HT2 receptors is an interest- jority of active St John’s wort compounds in a wideing pharmacological finding but does not explain the range of receptor assays to provide a broad in vitro

pharmacological profile.[28]mode of its antidepressant action.

In conclusion, the results presented in sections 3.6.1 St John’s Wort ExtractIn prior studies, a commercially available crude3.5.1 and 3.5.2 indicate that the actions of antide-

extract from the fresh flowers and buds of St John’spressants in general and St John’s wort specificallywort was tested in assays in a battery of 39 non-are probably not mediated only by simple up- orcloned receptors and two enzyme systems.[36] The

downregulation of monoamines and their receptors.crude extract of St John’s wort had significant affin-

Data on β-adrenoceptor function and serotonin ity for adenosine (subtype unspecified), GABAAreceptor regulation are summarised in table III. and GABAB, serotonin (subtype unspecified), ben-

Table III. Effects of St John’s wort extract and various compounds on β-adrenoceptors and serotonin receptors

Reference Compound After 2 weeks After 8 weeks After 26 weeks

β-Adrenoceptors

Muller et al.[37] Methanolic St John’s wort extract ↓ ND ND

Imipramine ↓ ND ND

Muller et al.[48] CO2 extract ↓ ND ND

Imipramine ↓ ND ND

Serotonin receptors

Muller et al.[37] Methanolic St John’s wort extract ↑ (5-HT2) ND ND

Muller et al.[48] Methanolic St John’s wort extract ↑ (5-HT2) ND ND

CO2 extract (↓) [5-HT2] ND ND

Teufel-Mayer and Gleitz[83] Methanolic St John’s wort extract ND ND ↑ (5-HT1A)

Methanolic St John’s wort extract ND ND ↑ (5-HT2A)

Butterweck et al.[27] Methanolic St John’s wort extract – ↓ (5-HT1A) ND

Hypericin – ↓ (5-HT1A) ND

Imipramine – ↓ (5-HT1A) NDND = not determined; – indicates no effect; ↓ indicates receptor downregulation; (↓) indicates slight receptor downregulation; ↑ indicatesreceptor upregulation.



zodiazepine and inositol triphosphate (IP3) receptors and δ-opioid receptors was found to be inhibited byand MAO-A and -B. In addition, Simmen and col- hyperforin and hypericin in the submicro- andleagues[96,97] showed that an ethanolic extract inhib- micromolar range.[97] This result could not be con-ited ligand binding to the GABAA receptor. firmed by Butterweck et al.,[28] who found thatBaureithel et al.[100] also showed that St John’s wort neither hyperforin nor hypericin affected opioid re-extracts inhibited [3H]flumazenil binding to ben- ceptor binding. In agreement with Butterweck etzodiazepine binding sites on the GABAA receptor. al.,[28] Raffa[53] also found that hypericin had no

affinity for opioid receptors.3.6.2 Individual Compounds The most potent effect on GABA receptors wasIsolated compounds of St John’s wort were seen for amentoflavone, which inhibited binding of

screened in several studies for activity at G-protein- ligands to the benzodiazepine receptor.[28] In thecoupled receptors including serotonin, adrenergic, same study, the flavonoids as well as hyperforinopioid, histamine, metabotropic glutamate and mus- exhibited only a weak affinity for GABA receptors,carinic acetylcholine receptors, ligand-gated ion and the hypericins remained inactive. Similar resultschannels including GABAA/B receptors, and various are reported by Gobbi et al.:[98] [3H]muscimol bind-neurotransmitter transporters. The data are contra- ing was not inhibited by hypericin, hyperforin ordictory; some of the investigated substances caused biapigenin.an unanticipated binding inhibition in several recep-

Data on the effects of hypericin in various recep-tor assays, whereas the same substances were inac-tor-screening models are contradictory: Raffa[53]

tive in the same receptor systems in other studies.found that hypericin had no affinity for traditionalIn a recent study, the phloroglucinol derivativemonoamine receptors or adrenergic, GABA, adeno-hyperforin showed considerable affinity for dop-sine or benzodiazepine receptors. The naphthodian-amine D1 and D5 receptors, whereas the biflavonoidthrone had modest affinity for muscarinic choliner-amentoflavone showed affinity for D3 receptors.[28]

gic receptors (subtype not measured) and similarNielsen et al.[101] and Baureithel et al.[100] focusedaffinity for σ-receptors. Gobbi et al.[98] showed thattheir investigations on the biflavone amentoflavone,hypericin inhibited ligand binding to NPY1, NPY2which bound to the brain benzodiazepine receptorsand σ-receptors. The authors found that these inhibi-with an affinity comparable to diazepam. This resulttory effects were light dependent, because they de-was recently confirmed by Butterweck and col-creased or disappeared when binding assays wereleagues.[28] Whereas the flavonoids hyperoside, iso-carried out in the dark.quercitrin, quercitrin, quercetin and miquelianin as

Cott[36] tested hypericin in a battery of 39 in vitrowell as hyperforin exhibited a weak effect on GABAreceptor assays, and hypericin showed affinity onlyreceptors, the hypericins remained inactive. A simi-for the NMDA receptor (Ki approximately 1 µmol/lar result was reported by Gobbi et al.,[98] who foundL). This result could not be confirmed by But-that [3H]flumazenil binding to benzodiazepine re-terweck et al.[28] In the same study, hypericin show-ceptors was inhibited by biapigenin but not byed high affinity for the D3 receptor (Ki = 34.5 nmol/hyperforin or hypericin.L) and negligible affinities (Ki >1000 nmol/L) forIn the study by Butterweck et al.,[28]

nearly all other tested receptors and transporters.amentoflavone had remarkable affinity for the δ-Interestingly, the affinity of hypericin for the D3opioid receptor subtype in the nanomolar range (Kireceptor subtype was much higher than that of the= 36.5 nmol/L). The effects of St John’s wort ex-atypical antipsychotic clozapine (Ki = 372.3 nmol/tracts and isolated compounds on opioid receptorsL).have been studied recently by Simmen et al.[96,97]

Binding of [3H]naloxone to µ- and κ-opioid recep- With regard to the contradictory effects of hyper-tors was inhibited in the presence of St John’s wort icin in various test models, it should be pointed outextracts. Most recently, binding of ligands to µ-, κ- that the pharmacological evaluation of hypericin in


552B

utterweck

A

dis D

ata

Info

rma

tion

BV 2003. A

ll righ

ts rese

rved

.C

NS D

rug

s 2003; 17 (8)

Table IV. In vitro effects of St John’s wort extracts and isolated compounds on various receptorsa

Receptor subtype St John’s CO2 Hypericin Pseudo- Hyper- Amento- Biapig- Hypero- Isoquer- Quercitrin Rutin Quercetin Miquelianinwort extract hypericin forin flavone enin side citrinextract

Serotonin

5-HT1A –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

5-HT1Dα –[28] –[28] –[28] +[28] –[28] –[28] –[28] –[28] –[28] –[28]

5-HT1Dβ –[96,97] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

5-HT2C –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

5-HT3 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

5-HT5a –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

5-HT6 –[98] –[98] +[96,97]; +[96,97]; +[96,97]; –[28] –[98] –[28,96,97] –[28] –[28] –[28] –[28] –[28]

–[28,98] –[28] –[28,98]

5-HT7 –[98] –[98] –[28,96-98] +[96,97]; +[96,97]; –[28] –[98] –[28,96,97] –[28] –[28] –[28] –[28] –[28]

–[28] –[28,98]

Dopamine

D1 –[36] –[28,53] –[28] +[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

D2 –[36] –[28,53] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

D3 +[28] +[28] –[28] +[28] –[28] –[28] –[28] –[28] –[28] –[28]

D4 +[28] +[28] –[28] –[28] –[28] –[28] –[28] –[28] +[28] –[28]

D5 –[28] –[28] +[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

Opioid

µ –[96,97] +[96,97]; +[96,97]; +[96,97]; –[28,96] –[97] –[28,97,99] –[28,96,97] –[28,96,97] –[28,96,97] –[28,96,97] –[28]

–[28,53] –[28] –[28]

δ –[96,97] +[96,97]; +[96,97]; +[96,97]; +[28] –[97] –[28,96,97] –[28] –[28] –[28] –[28] –[28]

–[28,53] –[28] –[28]

κ –[96,97] +[96,97]; +[96,97]; +[96,97]; –[28] –[97] –[28,96,97] –[28,96,97] –[28] –[28,96,97] –[28] –[28]

–[28,53] –[28] –[28]

Adrenergic

α2A –[28] –[28] –[28] +[28] –[28] –[28] –[28] –[28] –[28]

α2B –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

α2C –[28] –[28] –[28] –[28] –[28] –[28] +[28] –[28] +[28]

β1 –[36] –[53]; +[28] –[28]

β2 +[28] –[28]

α1A –[36] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

Continued next page

Mechanism

of Action of St John’s W

ort553

A

dis D

ata

Info

rma

tion

BV 2003. A

ll righ

ts rese

rved

.C

NS D

rug

s 2003; 17 (8)

Table IV. Contd

Receptor subtype St John’s CO2 Hypericin Pseudo- Hyper- Amento- Biapig- Hypero- Isoquer- Quercitrin Rutin Quercetin Miquelianinwort extract hypericin forin flavone enin side citrinextract

α1B –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

Acetylcholine

M1 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

M2 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] +[28]

M3 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

M4 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

M5 –[28] –[28] –[28] –[28] –[28] –[28] +[28] –[28] +[28]

GABA

GABAA +[36,96-98] –[98] –[28,36,53,98] –[28,98] –[28] –[98] –[28] –[28] –[28] –[28] –[28] –[28]

GABAB +[36] –[36]

Benzodiazepine +[36,100]; –[98] –[28,36,53,98,100] –[28] –[28,98] +[28,100,101] +[98]; –[28] –[28] –[28,100] –[28,100] –[28,100] –[28]

–[98] –[100]

Glutamate

NMDA –[36] –[28,53]; –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

+[36]

Phencyclidine –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

Histamine

H1 –[96,97] –[53] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

H2 –[96,97]

Peptides

V1 –[28,53] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

V2 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

V3 –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28] –[28]

NPY1,2 –[98] –[98] +[98]; –[98] –[98]

–[53]

NK1 –[53]

CRF1 +[97] +[97] –[97]

Estrogen-α –[96,97] +[96,97] –[97] –[98]

σ –[98] –[98] +[53,98] –[98]

a This table presents an overview of receptor binding studies but does not consider quantitative differences, because the original data from different authors had a different ‘cut-off’ for activity.

CRF = corticotrophin release factor; M = muscarinic; NK = neurokinin; NPY = neuropeptide Y; V = vasopressin; + indicates affinity for the receptor; – indicates no affinity for thereceptor.

554 Butterweck

most in vitro and in vivo studies is hampered by its tency and their ability to reduce immobility time inpoor solubility in most aqueous solutions. This im- the FST.[107] The specificity of this test for antide-portant fact will be discussed in more detail in pressants has been questioned because a large num-section 4. ber of nonantidepressants also reduce immobility.

Taken together, what relevance do the sum- False positive results have been reported for stimu-marised in vitro receptor binding findings have for lants, anticholinergics, antihistamines, pentobarbitalunderstanding the mode of action of St John’s wort? (phenobarbitone) and opiates.[107-111] It has beenClearly, because related compounds interact with found that whereas response to antidepressants per-various receptors, it can be hypothesised that addi- sists with long-term treatment, it disappeared withtive or synergistic effects induced by distinct com- long-term administration of antihistamines.[112] Thepounds at multiple targets might be responsible for effects of stimulants such as caffeine or ampheta-the antidepressant efficacy of St John’s wort. How- mine can be excluded by additional open field tests,ever, the concentration of compounds required for in which these agents show marked increases inhalf-maximal receptor occupancy in all mentioned locomotor activity.[113] Thus, the FST, when validat-studies appears to be quite high. Circulating concen- ed, appears particularly useful for the screening oftrations that are within two orders of magnitude of antidepressant drugs.[114]

the Ki values may be sufficient for subtle modula- In the FST, extracts of St John’s wort have con-tion of synaptic transmission.[102,103] A further limi- sistently been shown to decrease immobility timetation is that studies of intrinsic activity have not dose dependently after single-dose as well as afterbeen done. In addition, pharmacokinetic data for repeated treatment[20,22,26,29,73,115-117] without chang-relevant substances are needed. In vivo studies are

ing locomotor activity.[71,73] In all studies, the effi-necessary to establish the pharmacological rele-

cacy of St John’s wort extract was comparable tovance of these in vitro findings for therapeutic usage

that of standard antidepressants. The results in theof St John’s wort.

FST after repeated treatment and after additionalopen field tests confirm a specific antidepressant4. Behavioural Pharmacologyeffect of the plant extract.in Animals

During a bioassay-guided fractionation of aThe potential antidepressant activity of St John’s methanolic St John’s wort extract, two main frac-

wort was confirmed in a number of animal models tions were obtained that significantly decreased im-of depression. mobility time in the FST.[20] These fractions were

Human psychiatric disorders cannot be induced characterised by quantitative HPLC as being rich inin animals, so animal models cannot be claimed to hypericin, pseudohypericin and procyanidins (frac-reproduce human psychopathology. Instead, they tion IIIc) and flavonoids (fraction II).[20] Furtherare intended to induce changes that are sensitive to investigations focused on isolated compounds fromtherapeutic agents in a manner predictive of their these fractions. Interestingly, pure hypericin andeffects in humans.[104] In this review, two animal pseudohypericin did not reduce immobility timemodels are of particular interest: (i) the forced after single-dose pretreatment at doses comparableswimming test (FST) [‘behavioural despair’];[105]

to the entire extract; only the exceptionally highand (ii) the learned-helplessness paradigm.[106]

dose of hypericin 0.23 mg/kg (orally) was signifi-cantly active, whereas pseudohypericin indicated

4.1 Forced Swimming Testsome nonsignificant activity at about 0.5 mg/kg(orally).[26] When the fraction of procyanidins,The FST was developed by Porsolt et al.[105] Thewhich was itself not active in the FST, was recom-authors described procedures for simulating a statebined with the naphthodianthrones, hypericin wasof ‘depression’ in rats. There is, in fact, a significant

correlation for antidepressants between clinical po- significantly active at 0.028 mg/kg (orally) and



pseudohypericin was active at 0.166 mg/kg (orally). Chatterjee and coworkers[22] have recently point-ed to hyperforin as the major non-nitrogenous com-The procyanidin fraction as well as the pure procy-pound of St John’s wort. Oral administration of ananidins B2 and C1 increased the water solubility ofethanolic extract (hyperforin content 4.5%; doseboth hypericins, which may lead to a better bioavail-range 50–300 mg/kg) and a supercritical CO2 ex-ability.[26] However, after repeated treatment (14tract (hyperforin content 38.8%; dose range 5–30days), hypericin was active in the FST without sup-mg/kg) dose dependently reduced immobility timeplementation of procyanidin B2,[74] indicating thatof rats in the FST. Based on these results, the authorswhen hypericin is administered repeatedly, the en-conclude that “hyperforin is the major active com-dogenous level will reach pharmacodynamic con-pound of the extract”.[22] Unfortunately, purecentrations without a solubiliser. These interestinghyperforin was not tested in this study; theoreticalpharmacokinetic properties could also explain whydose-effect curves of hyperforin were calculated onhypericin was inactive in most in vitro experiments:its concentration in the tested extracts. These mathe-

the biological evaluation of hypericin in variousmatical calculated results nicely show that the (theo-

tests is hampered by its poor water solubility, which retical) activity of hyperforin in the FST correlatescan be increased with the presence of various procy- well with its concentration in the different extracts.anidins (see also section 3.6). Since the pure compound was not tested in the FST

The bioassay-guided fractionation yielded a frac- in that study, it cannot be excluded that the effects oftion that was mainly characterised by its high num- the CO2 extract are due to effects of so-far untestedber of flavonoids. The fraction significantly reduced compounds (e.g. adhyperforin).immobility time in the FST, and the effect was Taken together, the data indicate that the FST hascomparable to that of imipramine.[20] The flavonoid been proven to be a reliable animal model for the

evaluation of substances with antidepressant activi-fraction was further purified and some flavonol gly-ty. Extracts of St John’s wort (methanolic, ethanoliccosides such as hyperoside, quercitrin, isoquercitrinand CO2 extracts) as well as pure compounds suchand miquelianin, as well as the flavone glycosideas hypericin and several flavonoids have been prov-astilbin, were isolated[21] and tested for activity inen to be active in this model, with effects compar-the FST at doses comparable to their amounts pre-able to standard antidepressants.sent in the crude drug material.[29] Except for quer-

citrin and astilbin, all flavonoids (hyperoside, iso-4.2 Learned-Helplessness Paradigmquercitrin and miquelianin) were significantly active

in the FST after single-dose as well as after repeated The learned-helplessness paradigm was original-treatment (oral administration). In addition, no in- ly described by Overmier and Seligman[106] in dogs,crease of locomotor activity was observed, indicat- and it was then extended to a large number of othering that their effects in the FST were specific. The species including rats and humans.[118] According toaglycone quercetin did not show activity in the FST this paradigm, exposure to uncontrollable stress pro-when tested at 0.4 mg/kg (orally), a dose analogous duces performance deficits in subsequent learningto hyperoside. The data obtained by Butterweck et tasks that are not seen in subjects exposed to theal.[29] indicate that certain flavonol-3-O-glycosides identical stressor but able to control it.[118,119]

are active constituents of the St John’s wort extract. Learned helplessness could be reversed by sub-The activity seems to be bound to the sugar moiety chronic treatment (4–7 days) with a variety of anti-of the aglycone quercetin, in that its glucoside, ga- depressants including TCAs, MAO inhibitors andlactoside and glucuronide are active compounds. It atypical antidepressants and with electroconvulsive

shocks.[120-123]may be that these particular glycosides are absorbedfrom the intestine, whereas others are not. Addition- The learned-helplessness paradigm responds to aal studies are necessary to verify this hypothesis. wide range of clinically effective treatments, and


556 Butterweck

there are no false negatives. However, the effects of tered intraperitoneally is several times more potentthan the total hydroalcoholic extract when giveneither long-term anticholinergic treatment or anti-orally.histamines on learned helplessness have not been

Taken together, the results in the learned-help-assessed.[124] Also, the specificity of learned help-lessness paradigm show that different St John’s wortlessness as a model of depression has been question-extracts (hydroalcoholic, CO2) as well as pureed.[104] Therefore, there are serious doubts regardinghyperforin can reduce behavioural deficits inducedthe validity of learned helplessness as a specificby uncontrollable stress. How active other constitu-model of depression; its specificity is unclear, andents of St John’s wort in this model are needs to bethe model appears to predict patterns of symptomsfurther elucidated.that are not found to occur in depression.[125]

The data of the FST and the learned-helplessnessChatterjee and colleagues[99] were the first toparadigm are summarised in table V.investigate two different St John’s wort extracts in

the learned-helplessness paradigm. In this study, the5. Other Possible Modes of Actionescape failures in the group treated with ethanolic

extract (hyperforin content 4.5%) significantly and A relationship between endocrine disorders anddose dependently decreased after 150 and 300 mg/ depression has been noted for many years, and datakg/day (orally). Comparable effects were observed indicate functional disturbances of endocrinefor the CO2 extract (hyperforin content 38.8%) at systems in depression. The hypothalamic-pituitary-dosages of 15 and 30 mg/kg/day (orally). adrenal (HPA) axis, the hypothalamic-pituitary-thy-

roid (HPT) axis, the neuropituitary axis and theA similar study was performed by Gambrana etsomatotropic axis are particularly affected.[128-131] Aal.[126] The authors studied the effects of acommon biological alteration in patients with majorhydroalcoholic St John’s wort extract (hypericindepression is activation of the HPA axis, whichcontent 0.3%, hyperforin content unknown) on the

escape-deficit paradigm. The escape-deficit para-digm is a simplified version of the classical learned-helplessness syndrome.[106,118,123] St John’s wort ex-tract, given as a single dose at relatively high dosesof 1–1.5 g/kg (orally), significantly protected ratsfrom the consequences of exposure to acute una-voidable stress. Chronic stress induced a long-last-ing escape deficit that was completely reversed byrepeated pretreatment with imipramine, fluoxetineor St John’s wort extract. The authors further studiedthe efficacy of a CO2 extract (extract not furthercharacterised) in the acute escape deficit model aftersingle and repeated treatment.[127] A dose-responsecurve for the CO2 extract, administered orally acute-ly before stress exposure, showed that this prepara-tion was five times more potent than the total extract(extract not further characterised). Pure hyperforinwas also tested in the acute escape-deficit paradigmin a dose range of 12.5–75 mg/kg (intraperitonealinjection). Hyperforin 20, 50 and 75 mg/kg signifi-cantly counteracted the effects of acute stress. Theauthors concluded that hyperforin when adminis-

Table V. Effects of St John’s wort extract and various compoundsin the forced swimming test and learned-helplessness paradigm

Substance Forced swimming test Learnedhelplessness

Hydroalcoholic St John’s ↓[20,22,26,29,73,74,115-117] ↓[22,126,127]

wort extract

CO2 extract ↓[22] ↓[22,126,127]

Flavonoid fraction (II) ↓[20,29] ND

Hypericin + procyanidin ↓[20,26] NDfraction (IIIc)

Pure hypericin –[26] ND

Pure pseudohypericin –[26] ND

Procyanidin B2 –[26,74] ND

Pure hypericin + ↓[26,74] NDprocyanidin B2

Hyperoside ↓[29] ND

Quercitrin –[29] ND

Isoquercitrin ↓[29] ND

Miquelianin ↓[29] ND

Astilbin –[29] ND

Hyperforin ND ↓[127]

Imipramine ↓[20,22,26,29,73,74,115-117,126,127] ↓[22,126,127]

ND = not determined; ↓ indicates decrease of immobility time/escape deficit; – indicates no effect.



manifests itself as hypersecretion of adrenocortico- included in this study, so it is not clear if the ob-trophic hormone (ACTH) and cortisol and an abnor- served increase in plasma corticosterone level wasmal cortisol response to dexamethasone and cortico- due to the treatment or induced by stress during thetrophin-releasing hormone (CRH).[128,132,133] Corre- treatment procedure (intraperitoneal injection). Forspondingly, the hyperactivity of the HPA axis in example, similar increased corticosterone levelsdepressed patients can be corrected by effective were observed after acute immobilisation stress.[27]

therapy with antidepressant drugs.[134,135] Similarly,The study by Butterweck et al.[27] showed that thechronic stress in animals results in elevated ACTH

three compounds – imipramine, St John’s wort ex-and glucocorticoid levels in plasma and elevatedtract and hypericin – all produced similar delayedproduction of CRH in the parvocellular portion ofchanges in CRH gene transcription, suggesting athe hypothalamic paraventricular nucleusshared action on the centres that control HPA axis(PVN).[136-138]

activity. The data from Butterweck et al.[27] alsoAdministration of antidepressants influences the suggest that the delayed changes may be important

expression of genes coding for HPA-axis constitu-for the therapeutic efficacy of St John’s wort extract

ents. In an animal study designed to examine theand hypericin. This hypothesis is based on changes

association between long-term antidepressant ad-in mRNA expression in unstressed, eucortisolaemicministration and the possibly delayed alteration inrats.HPA axis activity, CRH mRNA in the PVN of rats

was shown to be decreased following long-term (8 The authors performed a second experiment toweeks) but not short-term (2 weeks) treatment with assess the effects of long-term administration ofimipramine, the prototypic TCA.[69] The same re- imipramine and St John’s wort extract on stress-sults were found with several other antidepressant induced changes in gene transcription in stress-re-drugs selected for their distinctly different primary sponsive circuits.[27] Depression-associated behav-pharmacological actions.[70] ioural symptoms can be observed in stress models,

and long-term but not acute administration of anti-Based on the results of Brady et al.,[69,70] But-depressants usually reverses these effects.[125,139,140]terweck et al.[27] performed a study to examine inHowever, Butterweck et al.[27] found that repeatedrats the effects of short-term (2 weeks) and long-immobilisation stress (2 hours daily for 7 days)term (8 weeks) administration of imipramine, Stincreased CRH mRNA levels in the PVN, pro-John’s wort extract and hypericin on the expressionopiomelanocortin (POMC) in the anterior pituitary,of genes that may be involved in the regulation ofglutamic acid decarboxylase (GAD 65/67) in thethe HPA axis. Imipramine (15 mg/kg orally), St

John’s wort (500 mg/kg orally) and hypericin (0.2 bed nucleus of the stria terminalis (BST), cyclicmg/kg orally) significantly decreased levels of CRH adenosine monophosphate response element bind-mRNA by 16–22% in the PVN when given daily for ing protein (CREB) in the hippocampus and tyrosine8 weeks but not for 2 weeks. Comparable to imip- hydroxylase in the locus coeruleus. It decreasedramine, both the extract and hypericin reduced plas- levels of mRNA for 5-HT1A receptors and brain-ma ACTH and corticosterone levels after 2 weeks of derived neurotrophic factor (BDNF) in the hippo-daily treatment. campus. Long-term pretreatment with either imip-

ramine or St John’s wort reduced to control levelsFranklin et al.[66] found increased plasma cortico-the stress-induced increases in gene transcription ofsterone levels in rats after single doses of St John’sGAD in the BST, CREB in the hippocampus andwort extract (200 mg/kg intraperitoneally), hyper-POMC in the pituitary. The stress-induced increasesicin (0.2 mg/kg intraperitoneally) and hyperforinin CRH mRNA in the PVN and tyrosine hydroxy-(9.3 mg/kg intraperitoneally). Median maximallase in the locus coeruleus were reduced by imip-plasma levels were observed 2 hours after injection.ramine but not by St John’s wort. The stress-inducedHowever, a vehicle-treated control group was not


558 Butterweck

decreases in BDNF and 5-HT1A mRNA were not pressant action of St John’s wort. Further evidenceis necessary to support this hypothesis.prevented by either drug.

The data of Butterweck et al.[27] show that (i) St6. Concluding RemarksJohn’s wort extract and hypericin have delayed ef-

fects on HPA axis control centres similar to those of Herbal medicines are complex mixtures of moreimipramine; and (ii) selective stress-induced than one active ingredient. Therefore, pharmacolog-changes in gene transcription in particular brain ical work is complicated by the fact that activeareas can be prevented by long-term treatment with compounds are often unknown. Further, synergisticeither the prototypic TCA imipramine or St John’s or antagonistic effects of the different compoundswort. However, imipramine appeared to be more cannot be excluded. Nevertheless, the example of Steffective in blocking stress effects on the HPA axis John’s wort offers convincing proof for the conceptthan the plant extract. that modern methods of pharmacological,

phytochemical, biochemical and molecular biologi-Further evidence that St John’s wort extract inter-cal research are effective in advancing the develop-acts with certain neuroendocrine pathways was re-ment of traditional herbal remedies with CNS activi-ported by Thiele and colleagues.[141] The authorsty. As a consequence of these efforts, today severalinvestigated in vitro the effect of St John’s wortcompounds from different structural groups andextract on lipopolysaccaride-stimulated whole bloodwith different mechanisms of action seem to beculture systems from five healthy volunteers andresponsible for the observed antidepressant efficacyfour depressed patients. The release of interleukinof St John’s wort. Based on recent research, it seems(IL)-6, IL-1β and tumour necrosis factor-α (TNFα)therefore likely that flavonoids, hyperforins andwas measured. Whereas St John’s wort extract mas-hypericins contribute to the antidepressant efficacysively suppressed IL-6 release from stimulatedof St John’s wort. Their presence in a sufficientblood cells, no effect was observed on IL-1β andamount in St John’s wort extracts should be guaran-TNFα. Similar results were obtained by Fiebich etteed to provide herbal products with a high pharma-al.[142] The authors showed that different extractsceutical quality.from St John’s wort dose dependently inhibited sub-

stance P-induced IL-6 synthesis in a human astro- Research on the active constituents of St John’scytoma cell line. Substance P is a neuropeptide that wort is far from finished. Furthermore, the mecha-is involved in the aetiology of neurogenic inflamma- nism of antidepressant action of the plant extract istion and, as discovered more recently, of depression still not fully understood. Our understanding of theand anxiety.[143] mode of action of St John’s wort is complicated by

the fact that the molecular basis of depression itselfBoth clinical and experimental studies indicateis still unclear. More than 50 years after the discov-that stress and depression are associated with in-ery of the first antidepressants, new findings andcreased circulating levels of cytokines such as IL-1βnew hypotheses are still being reported in an attemptand IL-6 and hyperactivity of the HPA axis. It isto provide more insight into the mode of action ofwell established that HPA-axis activation is elicitedthese substances. Considering the relatively shortby exogenous cytokine administration to rodents.time that St John’s wort has been under systematicThus, it was suggested that hypercortisolaemia ob-investigation, there has been much progress in ourserved in depressed patients might have resultedunderstanding of its CNS actions, but there is still afrom a hypersecretion of CRH induced by pro-lot to be done.inflammatory cytokines such as IL-1β or IL-6. It is

therefore not unreasonable to speculate that the StAcknowledgementsJohn’s wort-induced IL-6 inhibition may be respon-

sible for the decreased CRH expression. This aspect The author thanks Dr Miles Herkenham, Professor Drcould be of relevance for the mechanism of antide- Adolf Nahrstedt and Professor Dr Hilke Winterhoff for con-



19. Nahrstedt A, Butterweck V. Biologically active and other chem-tinued enthusiastic support, insightful discussions andical constituents of the herb of Hypericum perforatum L.thoughtful reading of this manuscript.Pharmacopsychiatry 1997; 30 Suppl. 2: 129-34

20. Butterweck V, Wall A, Lieflaender-Wulf U, et al. Effects of theNo sources of funding were used to assist in the prepara-total extract and fractions of Hypericum perforatum in animaltion of this manuscript. The author has no conflicts of interest assays for antidepressant activity. Pharmacopsychiatry 1997;

that are directly relevant to the content of this manuscript. 30 Suppl. 2: 117-2421. Jurgenliemk G, Nahrstedt A. Phenolic compounds from Hyper-

icum perforatum. Planta Med 2002; 68: 88-91References 22. Chatterjee SS, Bhattacharya S, Wonnemann M, et al.1. Rosenthal N. St John’s wort: the herbal way to feeling good. Hyperforin as a possible antidepressant component of Hyper-

New York: Harper Collins Publishers, 1998 icum extracts. Life Sci 1998; 63: 499-5102. Bombardelli E, Morazzoni P. Hypericum perforatum. Fitoter- 23. Singer A, Wonnemann M, Muller W. Hyperforin, a major

apia 1995; 66: 43-68 antidepressant constituent of St John’s wort, inhibits serotonin3. Lohse MJ, Muller-Oerlinghausen B. Psychopharmaka. In: uptake by elevating free intracellular Na+. J Pharmacol Exp

Schwabe U, Paffrath D, editors. Arzneiverordnungreport 2000. Ther 1999; 290: 1363-8Berlin: Springer Verlag, 2000: 581-7

24. Muller WE, Singer A, Wonnemann M. Hyperforin-antidepres-4. Schulz V. Incidence and clinical relevance of the interactions sant activity by a novel mechanism of action. Pharmacop-

and side effects of Hypericum preparations. Phytomedicine sychiatry 2001; 34 Suppl. 1: S98-S1022001; 8 (2): 152-60

25. Wonnemann M, Singer A, Siebert B, et al. Evaluation of5. National Center for Complementary and Alternative Medicine.

synaptosomal uptake inhibition of most relevant constituentsSt John’s wort fact sheet (publication Z-02). Bethesda: Nation-

of St John’s wort. Pharmacopsychiatry 2001; 41 Suppl. 1:al Institutes of Health, 1999

S148-S516. Brenner R, Azbel V, Madhusoodanan S, et al. Comparison of an

26. Butterweck V, Petereit F, Winterhoff H, et al. Solubilizedextract of Hypericum (LI 160) and sertraline in the treatment ofhypericin and pseudohypericin from Hypericum perforatumdepression: a double-blind, randomized pilot study. Clin Therexert antidepressant activity in the forced swimming test.2000; 22 (4): 411-9Planta Med 1998; 64: 291-4

7. Harrer G, Schulz V. Clinical investigation of the antidepressant27. Butterweck V, Winterhoff H, Herkenham M. St John’s wort,effectiveness of Hypericum. J Geriatr Psychiatry Neurol 1994;

hypericin, and imipramine: a comparative analysis of mRNA7 Suppl. 1: S6-8levels in brain areas involved in HPA axis control following8. Philipp M, Kohnen R, Hiller K. Hypericum extract versusshort-term and long-term administration in normal andimipramine or placebo in patients with moderate depression:stressed rats. Mol Psychiatry 2001; 6: 547-64randomised multicentre study of treatment for eight weeks.

28. Butterweck V, Nahrstedt A, Evans J, et al. In vitro receptorBMJ 1999; 319: 1534-8screening of pure constituents of St John’s wort reveals novel9. Schrader E. Equivalence of St John’s wort extract (Ze 117) andinteraction with a number of GPCR’s. Psychopharmacologyfluoxetine: a randomized, controlled study in mild-moderate2002; 162: 193-202depression. Int Clin Psychopharmacol 2000; 5 (2): 61-8

29. Butterweck V, Jurgenliemk G, Nahrstedt A, et al. Flavonoids10. Volz HP. Controlled clinical trials of Hypericum extracts infrom Hypericum perforatum show antidepressant activity indepressed patients: an overview. Pharmacopsychiatry 1997; 30the forced swimming test. Planta Med 2000; 66: 3-6Suppl. 2: 72-6

30. Calapai G, Crupi A, Firenzuoli F, et al. Effects of Hypericum11. Woelk H. Comparison of St John’s wort and imipramine forperforatum on levels of 5-hydroxytryptamine, noradrenalinetreating depression: randomised controlled trial. BMJ 2000;and dopamine in the cortex, diencephalon and brainstem of the321 (7260): 536-9rat. J Pharm Pharmacol 1999; 51: 723-812. Linde K, Ramirez G, Mulrow C. St John’s wort in depression-

31. Suzuki O, Katsumata Y, Oya M. Inhibition of monoamineand overview and meta-analysis of randomised clinical trials.oxidase by hypericin. Planta Med 1984; 50: 272-4BMJ 1996; 313: 253-8

32. Demisch L, Holzl J, Gollnik B. Identification of selective MAO-13. Wheatly D. Hypericum extract: potential in the treatment oftype-A inhibitors in Hypericum perforatum L. Pharmacop-depression. CNS Drugs 1998; 9: 431-40sychiatry 1989; 22: 194-614. Vorbach EU, Arnoldt KH, Hubner WD. Efficacy and tolerabili-

33. Sparenberg BL, Demisch J, Holzl J. Untersuchungen uber diety of St John’s wort extract LI 160 versus imipramine inantidepressiven wirkstoffe von johanniskraut. Pharm Ztg Wisspatients with severe depressive episodes according to ICD-10.1993; 138: 239-54Pharmacopsychiatry 1997; 30 Suppl. 2: 77-80

34. Bladt S, Wagner H. Inhibition of MAO by fractions and constit-15. Laakmann G, Schule C, Baghai T, et al. St John’s wort in milduents of Hypericum extract. J Geriatr Psychiatry 1994; 154:to moderate depression: the relevance of hyperforin for the125-34clinical efficacy. Pharmacopsychiatry 1998; 31 Suppl. 1: 54-9

35. Thiede HM, Walper A. Inhibition of MAO and COMT by16. Shelton RC, Keller MB, Gelenberg A, et al. Effectiveness of StHypericum extracts and hypericin. J Geriatr Psychiatry NeurolJohn’s wort in major depression. JAMA 2001; 285 (15):1994; 7: S54-S61978-86

36. Cott JM. In vitro binding and enzyme inhibition by Hypericum17. Effect of Hypericum perforatum (St John’s wort) in majorperforatum extract. Pharmacopsychiatry 1997; 30 Suppl. 2:depressive disorder. JAMA 2002; 287 (14): 1807-14108-1218. Nahrstedt A. Antidepressant constituents of Hypericum

perforatum. In: Chrubasik S, Roufogalis BD, editors. Herbal 37. Muller W, Rolli M, Schafer C, et al. Effects of Hypericummedicinal products for the treatment of pain. Lismore: South- extract (LI 160) in biochemical models of antidepressant activ-ern Cross University Press, 2000: 144-53 ity. Pharmacopsychiatry 1997; 30 Suppl. 2: 102-7


560 Butterweck

38. Cracchiolo C. Pharmacology of St John’s wort: botanical and 58. Krishtal O, Lozovaya N, Fisunov A, et al. Modulation of ionchemical aspects. Sci Rev Alt Med 1998; 2: 29-35 channels in rat neurons by the constituents of Hypericum

perforatum. Pharmacopsychiatry 2001; 34 Suppl. 1: S74-8239. Muller WE, Schafer C. Johanniskraut: in-vitro studie uberHypericum-extrakt (LI 160), hypericin und kampferol als an- 59. Chatterjee S, Filippov V, Lishko P, et al. Hyperforin attenuatestidepressiva. Dtsch Apoth Ztg 1996; 136: 17-24 various ionic conductance mechanisms in the isolated hippo-

campal neurons of the rat. Life Sci 1999; 65 (22): 2395-40540. Amara SG, Kuhar MJ. Neurotransmitter transporters: recentprogress. Annu Rev Neurosci 1993; 16: 73-93 60. Biber A, Fischer H, Romer A, et al. Oral bioavailability of

hyperforin from Hypericum extracts in rats and human volun-41. Shaskan EG, Snyder SH. Kinetics of serotonin accumulationteers. Pharmacopsychiatry 1998; 31 Suppl. 1: 36-43into slices from rat brain: relationship to catecholamine uptake.

J Pharmacol Exp Ther 1970; 175: 404-18 61. Yu PH. Effect of the Hypericum perforatum extract on serotoninturnover in the mouse brain. Pharmacopsychiatry 2000; 33:42. Snyder SH. Putative neurotransmitters in the brain: selective60-5neuronal uptake, subcellular localization, and interactions with

centrally acting drugs. Biol Psychiatry 1970; 2: 367-89 62. Serdarevic N, Eckert GP, Muller WE. The effects of extracts ofSt John’s wort and kava kava on brain neurotransmitter levels43. Fuller RW, Wong DT. Serotonin uptake and serotonin uptakein the mouse. Pharmacopsychiatry 2001; 34 Suppl. 1: S134-S6inhibition. Ann N Y Acad Sci 1990; 600: 68-78

44. Horn AS, Coyle JT, Snyder SH. Catecholamine uptake by 63. Fornal CA, Metzler CW, Mirescu C, et al. Effects of standard-synaptosomes from rat brain: structure-activity relationships of ized extracts of St John’s wort on the single-unit activity ofdrugs with differential effects on dopamine and norepine- serotonergic dorsal raphe neurons in awake cats: comparisonsphrine neurons. Mol Pharmacol 1971; 7: 66-80 with fluoxetine and sertraline. Neuropsychopharmacology

2001; 25 (6): 858-7045. Bourin M, Baker GB. The future of antidepressants. BiomedPharmacother 1996; 50: 7-12 64. Di Matteo V, Di Giovanni G, Di Mascio M, et al. Effect of acute

administration of Hypericum-CO2 extract on dopamine and46. Richelson E. Synaptic effects of antidepressants. J Clin Psycho-serotonin release in the rat central nervous system. Pharmacop-pharmacol 1996; 16 Suppl. 2: 1S-7Ssychiatry 2000; 33: 14-847. Gobbi M, Dalla Valle F, Ciapparelli C, et al. Hypericum

65. Rommelspacher H, Siemanowitz B, Mannel M. Acute andperforatum L. extract does not inhibit 5-HT transporter in ratchronic actions of a dry methanolic extract of Hypericumbrain cortex. Naunyn Schmiedebergs Arch Pharmacol 1999;perforatum and a hyperforin-rich extract on dopaminergic and360: 262-9serotonergic neurons in rat nucleus accumbens. Pharmacop-48. Muller WE, Singer A, Wonnemann M, et al. Hyperforin repre-sychiatry 2001; 34 Suppl. 1: S119-S26sents the neurotransmitter reuptake inhibiting constituent of

66. Franklin M, Chi JD, Mannel M, et al. Acute effects of LI 160Hypericum extract. Pharmacopsychiatry 1998; 31 Suppl. 1:(extract of Hypericum perforatum, St John’s wort) and two of16-21its constituents on neuroendocrine responses in the rat. J49. Neary JT, Whittemore SR, Bu Y, et al. Biochemical mechan-Psychopharmacol 2000; 14 (4): 360-3isms of action of Hypericum LI 160 in glial and neuronal cells:

67. Kaehler ST, Sinner C, Chatterjee SS, et al. Hyperforin enhancesinhibition of neurotransmitter uptake and stimulation of ex-the extracellular concentrations of catecholamines, serotonintracellular signal regulated protein kinase. Pharmacopsychia-and glutamate in the rat locus coeruleus. Neurosci Lett 1999;try 2001; 34 Suppl. 1: S103-S7262: 199-20250. Perovic S, Muller WE. Pharmacological profile of Hypericum

68. Philippu A. In vivo neurotransmitter release in the locusextract. Arzneimittel Forschung 1995; 45: 1145-8coeruleus-effects of hyperforin, inescapable shock and fear.51. Kientsch U, Burgi S, Ruedeberg C, et al. St John’s wort extractPharmacopsychiatry 2001; 34 Suppl. 1: S111-S5ZE 117 (Hypericum perforatum) inhibits norepinephrine and

69. Brady LS, Whitfield HJ, Fox RJ, et al. Long-term antidepres-serotonin uptake into rat brain slices and reduces β-adre-sant administration alters corticotropin-releasing hormone, ty-noceptor numbers on cultured rat brain cells. Pharmacop-rosine hydroxylase and mineralocorticoid receptor gene ex-sychiatry 2001; 34 Suppl. 1: S56-60pression in rat brain. J Clin Invest 1991; 87: 831-752. Jensen AG, Hansen SH, Nielsen EO. Adhyperforin as a contrib-

utor to the effect of Hypericum perforatum L. in biochemical 70. Brady LS, Gold PW, Herkenham M, et al. The antidepressantmodels of antidepressant activity. Life Sci 2001; 68: 1593-605 fluoxetine, idazoxan and phenelzine alter corticotropin-releas-

ing hormone and tyrosine hydroxylase mRNA levels in rat53. Raffa RB. Screen of receptor and uptake-site activity of hyper-brain: therapeutic implications. Brain Res 1992; 572: 117-25icin component of St John’s wort reveals S-receptor binding.

Life Sci 1998; 62 (16): 265-70 71. Butterweck V, Bockers T, Korte B, et al. Long-term effects ofSt John’s wort and hypericin on monoamine levels in rat54. Haslam E. Natural polyphenols (vegetable tannins) as drugs:hypothalamus and hippocampus. Brain Res 2002; 930: 21-9possible modes of action. J Nat Prod 1996; 59: 205-15

72. Franklin M, Cowen PJ. Researching the antidepressant actions55. Buchholzer ML, Dvorak C, Chatterjee SS, et al. Dual modula-of Hypericum perforatum (St John’s wort) in animals and man.tion of striatal acetylcholine release by hyperforin, a constitu-Pharmacopsychiatry 2001; 34 Suppl. 1: S29-37ent of St John’s wort. J Pharmacol Exp Ther 2002; 301 (2):

714-9 73. Winterhoff H, Butterweck V, Nahrstedt A, et al. Pharmako-logische untersuchungen zur antidepressiven wirkung von56. Wonnemann M, Singer A, Muller WE. Inhibition ofHypericum perforatum L. In: Loew D, Rietbrock N, editors.synaptosomal uptake of 3H-L-glutamate and 3H-GABA byPhytopharmaka in forschung und klinischer anwendung.hyperforin, a major constituent of St John’s wort: the role ofDarmstadt: Steinkopff Verlag, 1995: 39-56amiloride sensitive sodium conductive pathways. Neuropsy-

chopharmacology 2000; 23 (2): 188-97 74. Butterweck V, Korte B, Winterhoff H. Pharmacological and57. Fisunov A, Lozovaya N, Tsintsadze T, et al. Hyperforin modu- endocrine effects of Hypericum perforatum and hypericin after

lates gating of P-type Ca2+ current in cerebellar Purkinje repeated treatment. Pharmacospsychiatry 2001; 34 Suppl. 1:neurons. Eur J Physiol 2000; 440: 427-34 S2-7



75. Franklin M, Chi J, McGavin C, et al. Neuroendocrine evidence treatment with desipramine and fluoxetine alone and in combi-for dopaminergic actions of Hypericum extract (LI 160) in nation. J Neurochem 1994; 62: 2262-8healthy volunteers. Biol Psychiatry 1999; 46: 581-4 94. Hrdina PD, Vu TB. Chronic fluoxetine treatment upregulates

76. Tuomisto J, Mannisto P. Neurotransmitter regulation of anterior 5-HT uptake sites and 5-HT2 receptors in the rat brain: anpituitary hormones. Pharmacol Rev 1985; 37 (3): 249-311 autoradiographic study. Synapse 1993; 14: 324-31

77. Charney DS, Menkes DB, Heninger GR. Receptor sensitivity 95. Stahl S. 5HT1A receptors and pharmacotherapy: is serotoninand the mechanism of action of antidepressant treatment: receptor down-regulation linked to the mechanism of action ofimplications for the etiology and therapy of depression. Arch antidepressant drugs? Psychopharmacol Bull 1994; 30: 39-43Gen Psychiatry 1981; 38: 1160-80 96. Simmen U, Burkard W, Berger K, et al. Extracts and constitu-

78. Heninger GR, Charney DS. Mechanism of action of antidepres- ents of Hypericum perforatum inhibit the binding of varioussant treatments: implications for the etiology and treatment of ligands to recombinant receptors expressed with the Semlikidepressive disorders. In: Meltzer H, editor. Psychopharmaco- Forest virus system. J Recept Signal Transduct Res 1999; 19:logy: the third generation in progress. New York: Raven Press, 59-741987: 535-44 97. Simmen U, Higelin J, Berger-Buter K, et al. Neurochemical

79. Sulser F, Vetulani J, Mobley P. Mode of action of antidepressant studies with St John’s wort in vitro. Pharmacopsychiatry 2001;drugs. Biochem Pharmacol 1978; 27: 257-61 34 Suppl. 1: S137-S42

80. Banerjee SP, Kung LS, Riggi SJ, et al. Development of β- 98. Gobbi M, Moia M, Pirona L, et al. In vitro binding studies withadrenergic receptor subsensitivity by antidepressants. Nature two Hypericum perforatum extracts – hyperforin, hypericin1977; 268: 455-6 and biapigenin – on 5-HT6, 5-HT7, GABAA/benzodiazepine,

81. Vetulani J, Sulser F. Action of antidepressant treatments reduces sigma, NPY-Y1/Y2 receptors and dopamine transporters.reactivity of noradrenergic cAMP-generating system in limbic Pharmacopsychiatry 2001; 34 Suppl. 1: S45-S8forebrain. Nature 1975; 257: 495-6 99. Chatterjee SS, Noldner M, Koch E, et al. Antidepressant

82. Meltzer HY. Role of serotonin in depression. Ann N Y Acad Sci activity of Hypericum perforatum and hyperforin: the ne-1990; 600: 486-99 glected possibility. Pharmacopsychiatry 1998; 31 Suppl. 1:

83. Teufel-Mayer R, Gleitz J. Effects of long-term administration of 7-15Hypericum extracts on the affinity and density of the central 100. Baureithel KH, Buter KB, Engesser A, et al. Inhibition ofserotonergic 5-HT1A and 5-HT2A receptors. Pharmacop- benzodiazepine binding in vitro by amentoflavone, a constitu-sychiatry 1997; 30 Suppl. 2: 113-6 ent of various species of Hypericum. Pharm Acta Helv 1997;

84. Butterweck V. Beitrag zur pharmakologie und wirkstoff- 72 (3): 153-7findung von Hypericum perforatum L. Munster: Institut fur 101. Nielsen M, Frokjaer S, Braestrup C. High affinity of the natural-pharmazeutische Biologie und Phytochemie, Westfalische ly-occurring biflavonoid, amentoflavon, to brain ben-Wilhelms-Universitat, 1997 zodiazepine receptors in vitro. Biochem Pharmacol 1988; 37

85. Baker G, Greenshaw A. Effects of long-term administration of (17): 3285-7antidepressants and neuroleptics on receptors in the central 102. Kroeze WK, Roth BL. The molecular biology of serotoninnervous system. Cell Mol Neurobiol 1989; 9: 1-44 receptors: therapeutic implications for the interface of mood

86. Leonard BE. Mechanisms of action of antidepressants. CNS and psychosis. Biol Psychiatry 1998; 44: 1128-42Drugs 1995; 4 Suppl. 1: 1-12

103. Roth BL, Meltzer HY, Khan N. Binding of typical and atypical87. Leonard BE. The comparative pharmacological properties of antipsychotic drugs to multiple neurotransmitter receptors.

selective serotonin re-uptake inhibitors in animals. In: Feigh- Adv Pharmacol 1998; 42: 482-5ner JP, Boyer WF, editors. Selective serotonin re-uptake inhib-

104. Thiebot MH, Martin P, Puech AJ. Animal behavioural studies initors: advances in basic research and clinical practice. Chiches-the evaluation of antidepressant drugs. Br J Psychiatry Supplter: John Wiley and Sons, 1996: 35-621992; 160 (15): 44-50

88. Watanabe Y, Sakai R, McEwen B, et al. Stress and antidepres-105. Porsolt R, Le Pichon M, Jalfre M. Depression: a new animalsant effects on hippocampal and cortical 5-HT1A and 5-HT2

model sensitive to antidepressant treatments. Nature 1977;receptors and transport sites for serotonin. Brain Res 1993;266: 730-2615: 87-94

106. Overmier JB, Seligman MEP. Effects of inescapable shock upon89. Lopez J, Chalmers D, Little K, et al. Regulation of serotonin1A,subsequent escape and avoidance learning. J Comp Physiolglucocorticoid and mineralocorticoid receptor in rat brain andPsychol 1967; 78: 340-3human hippocampus: implications for the neurobiology of

107. Porsolt RD. Behavioural despair. In: Enna SJ, Malick JB,depression. Biol Psychiatry 1998; 43: 547-73Richelson E, editors. Antidepressants: neurochemical, behav-90. Briley M, Moret C. Neurobiological mechanisms involved inioural and clinical perspectives. New York: Raven Press, 1981:antidepressant therapies. Clin Neuropharmacol 1993; 16:121-39387-400

108. Betin C, De Feudis FV, Blavet N, et al. Further characterization91. Blier P, de Montigny C. Current advances and trends in theof the behavioural despair in mice: positive effects of convul-treatment of depression. Trends Pharmacol Sci 1994; 15:sants. Physiol Behav 1982; 28: 307-11220-6

109. Browne RG. Effects of antidepressants and anticholinergics in a92. Eison A, Yocca FD, Gianutsos G. Effect of chronic administra-mouse “behavioural despair” test. Eur J Pharmacol 1979; 58:tion of antidepressant drugs on 5-HT2 mediated behaviour in331-4the rat following noradrenergic or serotonergic denervation. J

Neural Trans 1991; 84: 19-32 110. Schlechter MD, Chance WT. Non-specificity of “behavioural93. Goodnough DB, Baker GB. 5-hydroxytryptamine 2 and β- despair” as an animal model of depression. Eur J Pharmacol

adrenergic receptor regulation in rat brain following chronic 1979; 60: 139-42


562 Butterweck

111. Wallach MD, Hedley LR. The effects of antihistamines in a 129. Holsboer F, Barden N. Antidepressants and hypothalamic-pitui-modified behavioural despair test. Commun Psychopharmacol tary-adrenocortical regulation. Endocr Rev 1996; 17: 187-2051979; 3: 35-9

130. Ansseau M. Hormonal disturbances in depression. In: Honig A,112. Kitada Y, Miyauchi T, Satoh A, et al. Effects of antidepressants Van Praag HM, editors. Depression: neurobiological, psycho-

in the rat forced swimming test. Eur J Pharmacol 1981; 72: pathological and therapeutic advances. Chichester: John Wiley145-52 & Sons, 1997: 235-50

113. Porsolt R, Anton G, Blavet N, et al. Behavioural despair in rats: 131. Vetulani J, Nalepa I. Antidepressants: past, present and future.a new model sensitive to antidepressant treatments. Eur J Eur J Pharmacol 2000; 405: 351-63Pharmacol 1978; 47: 379-91

132. Gold P, Goodwin F, Chrousos G. Clinical and biochemical114. Cryan JF, Markou A, Lucki I. Assessing antidepressant activity manifestations of depression: relation to the neurobiology of

in rodents: recent developments and future needs. Trends stress. N Engl J Med 1988; 319: 348-53Pharmacol Sci 2002; 23 (5): 238-45

133. Raadsher F, Hoogendijk W, Stam F, et al. Increased numbers of115. De Vry J, Maurel S, Schreiber R, et al. Comparison of Hyper- corticotropin-releasing hormone expressing neurons in the hy-

icum extracts with imipramine and fluoxetine in animal models pothalamic paraventricular nucleus of depressed patients.of depression and alcoholism. Eur Neuropsychopharmacol Neuroendocrinology 1994; 60: 436-441999; 9: 461-8

134. Barden N, Reul J, Holsboer F. Do antidepressants stabilize116. Ozturk Y. Testing the antidepressant effects of Hypericum spe- mood through actions on the hypothalamic-pituitary-adrenal

cies on animal models. Pharmacopsychiatry 1997; 30 Suppl. 2: axis? Trends Neurosci 1995; 18: 6-11125-8

135. Pepin M, Govindan M, Barden N. Increased glucocorticoid117. Panocka I, Perfumi M, Angeletti S, et al. Effects of Hypericumreceptor gene promoter activity after antidepressant treatment.perforatum extract on ethanol intake, and on behavioural de-Mol Pharmacol 1992; 41: 1016-22spair: a search for the neurochemical systems involved. Phar-

macol Biochem Behav 2000; 66 (1): 105-11 136. Herman J. In situ hybridization analysis of vasopressin genetranscription in the paraventricular and supraoptic nuclei of the118. Maier SF, Seligman MEP. Learned helplessness: theory andrat: regulation by stress and glucocorticoids. J Comp Neurolevidence. J Exp Psychol 1976; 1: 3-461995; 363: 15-27

119. Garber J, Miller WR, Seaman SF. Learned helpless, stress and137. Mamalaki E, Kvetnansky R, Brady L, et al. Repeated immobili-the depressive disorders. In: Depue RA, editor. The psychobi-

zation stress alters tyrosine hydroxylase, corticotropin-releas-ology of depressive disorders: implications for the effects ofing hormone and corticosteroid receptor messenger ribonucle-stress. New York: Academic Press, 1979: 335-63ic acid levels in rat brain. J Neuroendocrinol 1992; 4 (6): 690-5

120. Dorworthy TR, Overmeier JB. On “learned helplessness”: the138. Sawchenko P, Brown E, Chan R, et al. The paraventriculartherapeutic effects of electroconvulsive shock. Physiol Behav

nucleus of the hypothalamus and the functional neuroanatomy1977; 4: 355-8of visceromotor responses to stress. Prog Brain Res 1996; 107:

121. Martin P, Soubrie P, Simon P. The effect of monoamine oxidase 201-22inhibitors compared with classical tricyclic antidepressants on

139. Fuchs E, Kramer M, Hermes B, et al. Psychological stress inlearned helplessness paradigm. Prog Neuropsychopharmacoltree shrews: clomipramine counteracts behavioural and endo-Biol Psychiatry 1987; 11: 1-7crine changes. Pharmacol Biochem Behav 1996; 54 (1):

122. Martin P, Soubrie P, Puech AJ. Reversal of helpless behaviour 219-28by serotonin uptake blockers in rats. Psychopharmacology1990; 101: 403-7 140. Plaznik A, Palejko W, Stefanski R, et al. Open field behaviour

of rats reared in different social conditions: the effects of stress123. Sherman AD, Sacquinte JL, Petty F. Specificity of the learnedand imipramine. Pol J Pharmacol 1993; 45 (3): 243-52helplessness model of depression. Pharmacol Biochem Behav

1982; 16: 449-54 141. Thiele B, Brink I, Ploch M. Modulation of cytokine expressionby Hypericum extract. J Geriatr Psychiatry Neurol 1994; 7:124. Willner P. The validity of animal models of depression. Psycho-S60-S2pharmacology 1984; 83: 1-16

142. Fiebich BL, Hollig A, Lieb K. Inhibition of substance P-induced125. Willner P. Animal models of depression: an overview. Pharma-cytokine synthesis by St John’s wort extracts. Pharmacop-col Ther 1990; 45: 425-55sychiatry 2001; 34 Suppl. 1: S26-S8

126. Gambrana C, Ghiglieri O, Tolu P, et al. Efficacy of an Hyper-143. Maubach KA, Ruoniak NM, Kramer MS, et al. Novel strategiesicum perforatum (St John’s wort) extract in preventing and

for pharmacotherapy of depression. Curr Opin Chem Biolreverting a condition of escape deficit in rats. Neuropsycho-1999; 3: 481-8pharmacology 1999; 21 (2): 247-54

127. Gambrana C, Tolu PL, Masi F, et al. A study of the antidepres-sant activity of Hypericum perforatum on animal models.

Correspondence and offprints: Dr Veronika Butterweck, In-Pharmacopsychiatry 2001; 34 Suppl. 1: S42-S4stitute of Pharmacology and Toxicology, Univer-

128. Holsboer F, Spengler D, Heuser I. The role of corticotropin-sitatsklinikum Munster, Domagkstrasse 12, 48149 Munster,releasing hormone in the pathogenesis of Cushing’s disease,Germany.anorexia nervosa, alcoholism, affective disorders and demen-

tia. Prog Brain Res 1992; 93: 385-417 E-mail: [email protected]


Documents

Mechanism of Action of St John’s Wort in Depressionhomepages.gac.edu/~jwotton2/PSY385/mechanism in st. johns wort.pdf · Mechanism of Action of St John’s Wort 541 and, thus, it