Kava-kava in Modern Drug Research: Portrait of a Medicinal Plant

Kava-kava in Modern Drug Research: Portrait of a Medicinal Plant

Reference: Zeitschrift für Phytotherapie 1996; 17:180-95.

Summary: Recently, numerous reviews on kava have been published. Contrary to the ethnomedical and pragmatic-therapeutic oriented articles, the article presented here concentrates on the importance of isolated kava constituents as potential antiepileptics. This review analyzes the results of several doctoral theses which have not yet been published in commonly accessible journals. In rhizomes and stems of Piper methysticum, besides the familiar kava-pyrone and chalcone pigments, dimeric yangoin-derivatives are found, plus very small amounts of stigmastendion and an oxaporphinal-alkaloid (cepharadion A). The total kava extract as well as the isolated kava-pyrones have a protective effect against convulsions induced by poisons and electrical current. In phase II clinical trials, the kava extract and methysticin were effective in major clonic-tonic seizures, but exerted undesired side effects when applied long-term or in high doses, leading to discontinuation of the trial. Synthetic variations of methysticin, according to the results of pharmacological tests, appear to be unqualified as antiepileptics. Piperolide, isolated from Piper sanctum, is a variation of methysticin and at all dosages exerts an anticonvulsive effect. Also with respect to the duration of anticonvulsive effects, piperolide is superior to methysticin. Fadenolide, isolated from Piper fadyenii, represents a piperolide shortened by a C2-chain; however, pharmacological studies have not begun. Synthetic variations of dihydro-derivatives of fadyenolide have anticonvulsive properties. One of these variations has also proved to be as active as an antiepileptic in clinical trials. Kava-pyrones act as sodium channel blockers. Their antiepileptic effect, similar to that of phentoin and of other antiepileptics, could therefore be based on a decrease in the conductivity of certain cerebral areas. Recent investigations have shown that kava-pyrones at low concentrations bind to the histamine-H3 receptors. Because H3 antagonists are known to exert anticonvulsive effects, the addition of H3 antagonism should contribute to the anticonvulsive and antiepileptic effect of kava-pyrones and chemically related substances.

Until the fifties, almost all pharmaceutical research depended on medicinal plants as the source for new drugs. The first historical example is opium, which came to be used as an analgesic by way of morphine. A newer example is the rauwolfia root, which made the development of synthetic, mebeverine-type antispasmodics possible via reserpine. The plant kingdom was the source of new drugs. The method of screening traditionally used medicines is often considered ineffective today and many of the large pharmaceutical companies have given up their costly research into "natural programs." The example of kava, however, shows that persistent research can lead to the discovery of new active substances.

Today kava extracts are used therapeutically because of their anxiolytic effect. However, it is not the aim of this article to discuss the development of kava extract into an anxiolytic drug. The pharmacologically active constituents from the drug revealed an efficacy profile that resembles those of commonly used anticonvulsants, but without having any similarities in chemical makeup.

Over four decades ago kava extract was tested for its clinical usefulness in treating epilepsy. These trials were stopped, however, because there was no indication that kava was clinically superior to existing antiepileptic medicines. In a species botanically related to the kava plant, variants of the kava pyrones were found to have a more pronounced anticonvulsant effect. The chemical variation of the piperolide molecule facilitated the optimization of the efficacy profile and, in the end, led to the group of 4-methoxybutenolides as a group of new antiepileptics. The goal of this article is to describe this development in detail.

The Plants of Origin

Piper methysticum is a slow-growing perennial with a wreath-like disposition. The main stem is monopodial and the peripheral branches are sympodial. Generally, the main stem grows so that the lower side branches die and leave behind bumpy scars on the stem nodes. The fully grown plant looks like a giant bouquet, leafy and bushy on the top, with a bundle of ligneous stems at the base. The different cultivars vary in their disposition not inconsiderably. There are prostrate (very short internodes), normal (many single stems) and upright (only a few single stems with very long internodes) growth patterns. Plant height varies between 1 and 4 meters. The large leaves (13 to 28 cm x 10 to 22 cm) have a deep, heart-shaped base and 9 to 13 main nerves, and are pubescent on the underside. Due to oil cells the leaves appear to be dotted with holes when held against the light. Like the other "new world" Piper species, Piper methysticum is dioecious. The small, invisible male flowers (with no perianth) form a corncob-like inflorescence that is 3 to 9 cm long. Female plants are found only rarely, which is why there are very few descriptions of female flowers in flora literature. The flowers do exist,( 1) but don't develop into fruit. After pollination they fall off before fruit can develop. The high degree of polyploidization must contribute to infertility. The subterranean part of the plant consists of massive (2 to 10 kg in weight), branched, very lush rootstocks with many roots.

Piper wichmannii C. DC. is the wild form from which the different cultivars of Piper methysticum have developed.( 1) Piper wichmannii (synonyms: P. erectum C. DC., P. schlechteri C. DC. and P. arbuscula T release) is easy to confuse with P. methysticum because of its growth shape and morphological traits, but the inflorescences are longer and the species is fertile. The species is found mainly in Papua, New Guinea and the Solomon Islands.

Piper sanctum (Miqu.) Schlechtend. is native to Mexico, especially the provinces of Vera Cruz, Oaxaca, Morelos and San Luis Potosi. It grows into a 1 to 2 m high bunch and has easily broken twigs. The leaves are alternate, have short petioles, are large ( 20-25cm long, 14-18 cm wide) and heart-shaped at the base. The blossoms (inflorescences) are long and thin, which is why they are called "colas de rat¢n" (mouse tails) locally.( 2)

The Plant Drug

The kava rootstock, kava rhizome, usually consists of peeled, cut and dried Piper methysticum rootstock from which the roots have been removed. The drug smells mildly aromatic and strangely earthy. It tastes a little bitter and somewhat soapy. When the drug is chewed, saliva flow is stimulated and a long-lasting anesthesia of the tongue and gums occurs. An anatomical description can be found in reference 3. It is worth mentioning that the lipophilic kava pyrones are localized in the cells of the pith. While most of these cells are filled with starch, others contain a yellow "resin" that turns red when exposed to sulfuric acid (90%).

The portion of the plant used to manufacture the kava dry extract consists only in part of the rhizome. Lateral roots, dried sprout axes and kava peelings (strips of peeled, partly rolled pieces of rootstock bark that are brown on the exterior and yellow inside) make up a large portion of this commercial product.

Piper sanctum Root

This drug consists of dried Piper sanctum roots. No morphological descriptions of the entire or cut drug have been published. A batch of the drug brought to Germany consisted of dried, gray-brown root parts that ranged from 1.5 to over 5 cm thick. Cross-sections (magnified) clearly revealed the medullary ray construction of the wood. As seen under a microscope, the sections between the medullary rays are interrupted by noticeable cross-connections made up of parenchyma cells.( 4) These cells contain excretions which turn red. This leads to the conclusion that these are the localization points for the kava pyrones and piperolides.

Active Drug Constituents

A natural part of pharmacological research is to use chemical fractionation to look for a common property with the ultimate goal of isolating constituents that determine efficacy. Indeed, if at all possible, it's a test that is representative of the efficacy in humans. Practical tests for proving anxiolytic and centrally relaxing effects were first developed in connection with the discovery of mephenesine's "tranquilizing" effect by Frank Berger in 1946.( 5, 6) In 1874 the Berlin pharmacologist and toxicologist Louis Lewin (1850-1929) chose (for lack of other methods) the anesthetizing taste of the fractions as a common property in his search for effective constituents, possibly in analogy to the already discovered cocaine. In 1874, a monograph was published by L. Lewin on Piper methysticum. This study is still worth reading today as a source of ethnomedical and experimental-pharmacological information about this South Sea drug. "While chewing the root, one senses at first a spicy, mildly bitter taste, which evolves into a prickly, sharp, biting or peppery sensation, depending on the amount of the chewed substance...After the stinging sensation has lasted for about a minute, the tongue becomes sort of numb, and is sensitive to touch as the numbness decreases. This can last for many minutes and dwindles gradually until it's completely gone."( 7)

Lewin recognized that the anesthetizing constituent was highly lipophilic and that it could be extracted from the drug using petroleum ether. He described what was left after the extraction as a resin. "A drop from the point of a pin put on the tongue causes the abnormal sensations described above,...the sensation of numbness as well as the provable reduction of the feeling of the gums and all the other parts of the oral cavity that come in contact with the extract. A pin-sized dose of the resin put into a rabbit or guinea pig eye results in long-lasting, complete anesthesia of the cornea and conjunctiva."

The lipophilic "resin fraction" represents, from a chemical perspective, a mixture that, according to the usual phytochemical separation methods (adsorption chromatography, partition chromatography and fractionated crystallization( 8-11)), breaks down into crystallized single components. The dominant constituents fall into the category of kava pyrones. These are further categorized as either enolide pyrones (kavain and methysticin) or dienolide pyrones (yangonin). From the fractions that were soluble in lipid solutions many other substances were isolated, including:

- dimeric yangonin,( 8)

- 11-methoxy-yangonin,( 12, 13)

- tetrahydroyangonin,( 13)

- dimethoxydehydrokavain,( 14)

- sitosterol accompanied by stigmasten-4-dione-3,6,( 15)

- two chalkone derivatives described as flavokavins,( 16)

- cinnamic acid pyrrolidides,( 17) and

- an oxaporphine alkaloid, cepharadione A( 8) (see Figures 1 and 2).



Kava pyrones are found in Piper wichmanii( 18) and in Piper sanctum.( 19) In Piper sanctum the kava pyrones are accompanied by 5-branched lactones, the so-called piperolides (Figure 3).( 19, 20) The piperolides, chemically aryl-substituted cinnamylic butenolides, which are understood to be biosynthetic variants of the yangonin derivatives, coming about via ring reduction as a result of oxydative metabolism reactions.


One should remember that a medical substance very rarely has only one effect. As a rule, several effects come together to make up an efficacy profile. Additionally, the dose-response relationship in a comparison of active ingredients can be characteristic. The dose-efficacy curve can run flat or it can be biphasic, and show, in other words, a so-called "reverse effect."

For example, the kava pyrones have an anticonvulsive action in the higher dose range. In contrast, at lower dosages they increase the action of the convulsant poisons. Substances have to be compared in the entire dose range. In the two-phase course of the dose-response curve there is a difference, for example, in contrast to the benzodiazepines, with which the kava pyrones are often compared.( 21)


The following properties were tested:

- the superficial anesthetic effects on rabbit corneas (irritation using a flexible boar bristle) after application of the solution (0.25 ml) in the conjunctiva sack;

- the infiltration anesthetic effect with the guinea-pig-rash wheal test (intracutaneous injection of 0.25 ml of pyrones dissolved in peanut oil to test the stimuli response of the rash with needle irritation).( 22)

Kavain, dihydrokavain, methysticin and dihydromethysticin proved themselves to be clearly anesthetic in concentrations from 0.25% and up and cause complete anesthesia in concentrations of 1%. Mephenesin, procaine, benzocaine and cocaine were tested under the same conditions. The superficial anesthetic effect of the kava pyrones corresponded closely to that of cocaine, while benzocaine had a somewhat weaker effect. As to infiltration anesthesia, the kava pyrones had the same effect as benzocaine, but were three times weaker than procaine.( 23)


The following show that kavain-type kava pyrones (enolide derivatives) have genuine local anesthetic effects:

- the effect was concentration dependent in both forms of the anesthesia,

- the reversibility of the effect,

- the speed of the effect taking place, and

- the good tissue tolerance with repeated use.( 23)

With which medicines, that are not used therapeutically as local anesthetics, do the kava pyrones share the trait of being a local anesthetic as a component of their efficacy spectrum? Except for the central muscle relaxants, this is true for most of the antihistamines. First and foremost, the H( 1)-antihistamines have, along with their antihistamine properties, local anesthetic and antispasmodic, as well as sedative, effects.( 24)

Next it can be asked if the kava pyrones have other properties in common with antihistamines and local anesthetics. The efficacy spectrum of local anesthetics includes, among others, endoanesthetic, cerebral-protective, antifibrillatory and anticonvulsant effects; in addition the properties of the H( 1)-antihistamines include antispasmodic and sedative effects.


Endoanesthesia is understood to be the reduction of sensitivity of receptors of internal organs or tissue by a local anesthetic carried through the bloodstream. Proving endoanesthetic effects is usually done using electrophysiological research methods. The suppression of the Bezold-Jarisch reflex, a "vagus degeneration reflex," is used as a pharmacological proof method. It is characterized by slowed heart rate, distended vessels and lowered blood pressure. The receptors that trigger the Bezold-Jarisch reflex are found in the ventricular cardiac musculature. They are stimulated through deficient myocardial circulation, and by chemical substances, especially by veratrine. Local anesthetics as well as antihistamines are capable of preventing the veratrine-conditioned Bezold-Jarisch reflex.( 25)

On conscious rabbits the reflex is triggered through an i.v. injection of 80 æg veratrine sulfate/kg. It can be shown that the Bezold-Jarisch reflex can be eliminated by the kava pyrones kavain, dihydrokavain, methysticin and dihydromethysticin. After dosing with 5 to 20 mg pyrones/kg, the symptoms of a marked lowering of the heart rate and the continued reduction of the diastolic and systolic blood pressure fail to appear.


Substances with anticonvulsant effects are all potential antiepileptic drugs. Research is being done to see if the substances can suppress experimentally induced convulsions. Test models used include electrically induced spasms, chemically triggered spasms and the so-called kindling model. Substances with anticonvulsant effects can be differentiated by the fact that they influence clonic and tonic spasm phases differently. Tonic spasms consist of one continual muscle contraction, while clonic spasms come as quick convulsions of the antagonistic muscles following in rapid succession. This difference is important: the kava pyrones inhibit tonic spasm equivalents, albeit only in the higher dose range. In contrast, they intensify clonic spasms.

Figure 4 (taken from a publication by R. Kretzschmar and H. J. Meyer) shows how the tonic spasm components are influenced by kava pyrones after introduction of convulsant substances.( 26) While phenobarbital and phenytoin (diphenylhydantoin), two known antiepileptic drugs, cause, from the smallest dose up (over the total dose interval), an increasing inhibition of the tonic spasm phases, small doses of the kava pyrones dihydrokavain and dihydromethysticin encourage tonic pentetrazol spasms. The kava constituents kavain and methysticin show a biphasic response curve. These pyrones behave similarly toward the convulsants picrotoxin and bemegrid,( 23, 26) but not toward strychnine,( 26-29) whose special role is described in its own section below. Piperolide (which has an anticonvulsant effect for every dose level) doesn't exhibit biphasic characteristics toward convulsants. In comparison to the kavains and methysticins, it has a longer lasting effect.( 30)

The local anesthetic procaine shows a biphasic dose-response curve much the same as that of the kava pyrones. This similarity to procaine, benzocaine and cocaine indicates that local anesthetics and kava pyrones could have a similar effective mechanism.( 23, 31)


Fibrillation is understood to be irregular and asynchronic contractions of muscle, for example atrial and ventricular fibrillation. It is known that local anesthetics have an antifibrillatory effect that goes along with their primary function as anesthetics. For example, procainamide is used therapeutically as an antiarrhythmia drug. It has been known for over 30 years that enolide-type kava pyrones exhibit an antifibrillatory effect.( 32) The following test methods were used:

- electric stimulation of the ventricle of a guinea pig in situ;

- ventricular arrhythmia and extrasystole caused by i.v. injection of k-strophanthin or noradrenalin on conscious rabbits.

In all three research protocols the kava pyrones (20 mg/kg dose) were proven effective. For example, a dose of 150 æg strophanthin/kg (i.v., spread out over one hour) on conscious rabbits led to the loss of sinus rhythm and to ventricular extrasystole. Through gradual injections of 20 mg/kg dihydromethysticin this situation was immediately reversed. The simultaneous heavy dyspnea and cyanosis were alleviated immediately after dosing with dihydromethysticin.( 32)

The kavains and methysticins behave much like local anesthetics in their effect on the spontaneous activity and electrical sensitivity of isolated heart preparations (frog heart, guinea pig atria). In concentrations of 1 x 10(-5) g/ml they work negatively inotropically like cocaine, and in higher doses negatively chromotropically and negatively bathmotropically.( 32)


Lack of oxygen in the brain leads (independent of its genesis) to a series of disturbances in cell membranes and subcellular structures, especially of the mitochondria and further resulting in displacement of ions and changes in metabolic processes. Numerous animal models exist for measuring the individual course parameters of this phenomenon.( 33, 34) Substances which inhibit these disruptive processes are designated as having a cerebro- or neuroprotective effect. The following are included in the many substances that are considered cerebroprotective:

- most anticonvulsants, especially memantine, diphenylhydantoin (phenytoin) and carbamazepin;( 35)

- local anesthetics like lidocaine and procaine.( 36)

The question can then be asked, do the kava pyrones share a neuroprotective effect with the above substances?


In an experimental ischemia model on mice,( 34) the artificially induced brain infarction stayed significantly smaller under the influence of kava dry extract and after a dosage of methysticin and dihydromethysticin in the area of infarction. Specifically, a dose of the kava extract, standardized to a pyrone content of 70% (150 mg/kg p.o. one hour before occlusion of the median cerebral artery) reduced the size of the cerebral infarction in mice (p < 0.05, placebo comparison), the same as the anticonvulsant memantine (20 mg/kg i.p.). Kavain, dihydrokavain (respectively 30 and 60 mg/kg i.p.) and yangonin (70 and 140 mg/kg i.p.) didn't show any effects. In rats, the extract works as an anti-ischemic drug in the same pharmacological model with the same dosage and application type.( 37)


The aqueous drug extract, as well as the isolated kava pyrones including the yangonins, have an antispasmodic effect on isolated guinea pig ileum.( 38) Depending on the dosage, the contractions of the isolated guinea pig ileum caused by acetylcholine, histamine, nicotine and 5-hydroxytryptamine were inhibited. The effective concentration range is 10(-6) to 10(-5) g/l and corresponds to the narrow concentration range of the classic anti-spasmodic papaverine. All pyrones most readily inhibited the 5-hydroxytryptamine and nicotine spasms, whose effect is connected to nervous structures. The pyrones behave like the local anesthetics benzocaine, cocaine and procainamide in relation to the different spasmodics.


The obvious great similarity of the effective processes of anticonvulsants, local anesthetics and the kava pyrones leads one to the assumption that there is a common effective mechanism on the molecular level. The molecular effective mechanism of local anesthetics exists in their binding to open Na+ channels in the nerve membrane. The induced channel closings are often only a few milliseconds long, but very frequent. They disturb the single channel flow in many very short spurts and make the Na(+) flow ineffective.( 42) For a number of anticonvulsants it has been recognized that there is a blockade of Na(+) channels similar to that of tetradotoxin and a resulting lessening of the nerve conductivity that is the molecular basis for efficacy.( 43) The cerebroprotective effects of local anesthetic substances of the lidocaine type can also be connected to the Na(+) channel blockade.( 36) With cerebral oxygen deficiency there is an imbalance between supply and demand of metabolic substances. The energy deficiency in the initial stage of a brain edema can be remedied in two ways:

- improved circulation;

- reducing the tissue's need for energy.

The slowing of the metabolism is managed by hypothermia( 44) or by means of barbiturates.( 45) Another very effective method to reduce the need for energy can be found in the Na(+) blockade, since a large portion of the energy used in the brain is used to secure the stability of the membrane potential. Even with barbiturate-induced anesthesia, the brain uses 50% of the supplied energy to maintain the potassium and sodium flow through the membranes. The search for appropriate Na+ channel blockers is a highly topical area today in developing effective medicines for cerebral function disorders.( 36)

It can be shown in a clear way that the kava pyrones (at least kavain) are sodium channel blockers.( 46) In the range of 10 to 40 æM, kavain inhibits, depending on the dose, the flow (stimulated by veratrin) of sodium isolated synaptosomes of rats. (IC( 50)= 86æM). It has long been known from veratrine (protoveratrine) that the Na(+) concentrations increase through increasing the passive Na(+) flow into the cells.( 47)



A noticeable pharmacological property of kava pyrones is their antagonism toward strychnine. Animals pretreated with pyrones,( 26, 27, 48) or the Piper sanctum lactone piperolide( 30) have survived even repeated doses of fatal amounts of strychnine. The prevention of fatality is basically the result of preventing tonic extension spasms. As it was described for phenobarbital for the first time, the prevention of fatality is connected to a modification of how the seizure happens -- the typical extension spasms don't occur and are replaced with a long-lasting, rapid, sustained clonus of regular frequency.( 49)

The methysticins and kavains, and similarly, piperolide, behave differently depending on the amount of the strychnine dose. After dosage of 2 mg of strychnine/kg and premedication of, for example, 30 to 150 mg of methysticin/kg, the protected animals varied their behavior very little compared to normal animals despite the strychnine dose. It was even possible for them to eat as usual. Only with higher strychnine doses (3 to 4 mg/kg) did the dosage range (characterized by preventing typical strychnine symptoms) of the pyrones attach itself to a second antilethal effective dosage range, where all animals had a characteristic spasm syndrome (which they survived) with generalized rapid convulsions of regular frequency (see Table 1).

The effective mechanism of the kava pyrone antagonism toward strychnine is not known. It's similar to barbiturates insofar as no link exists between antistrychnine effects and reflex inhibiting properties, since both effects happen in different dose ranges.( 23) Above and beyond this, the receptor bond experiments showed that the kava pyrones don't bind to glycine (strychnine sensitive) receptors.( 29)


The following application areas are allowed for kava dry extract (standardized for kava pyrone content): nervous anxiety, tension and excitation states.( 50) The official monographs( 50) recommend as a daily dosage preparations or market-ready drugs with kava pyrones corresponding to 60 to 120 mg. In clinical studies (for an overview see reference 51) and in the clinical observations( 52, 53) the dose corresponded to 70 to 210 mg kava pyrones, that is, converted to a 70 kg human, 1 to 3 mg kava pyrones/kg. The psychosedative effect on humans at low doses is typical for this category of tranquilizers. One can then ask if this is representative for the typical effective profile for tranquilizers in animal experiments. The efficacy spectrum of tranquilizers is characterized by the following:( 54)

- anticonvulsant effect;

- inhibition of polysynaptic spinal reflexes;

- inhibition of conditioned reflexes;

- calming effect.

The anticonvulsant effects of kava pyrones and the piperolides have already been discussed. The spasm-inducing effect of the pyrones in the low dose range is atypical for tranquilizers.

A property of kava pyrones that they have in common with tranquilizers, and also with the central muscle relaxants, is the inhibition of polysynaptic spinal cord reflexes, which leads to muscular relaxation and ataxia. The electromyographically registering resistance during passive extension of a muscle in a conscious rabbit can, depending on the dose, be inhibited by intravenous injection of kava pyrones. Yangonin proved itself to be most effective binding at a median dose of 3.7 mg/kg. Piperolide with an inhibiting dose of around 40 mg/kg (a threshold dose near the lethal range) has the weakest bond. To compare: average inhibiting dose of mephenesine = 13.0 mg/kg and from diazepam = 0.2 mg/kg.( 28, 55, 56)

The muscle relaxing effect of pyrones can be traced to a centrally dispersed relaxing of the a- and g-spinal-motor system, which occurs (similarly as for mephenesine) by supraspinal fits. The inhibition of the tonic extension reflex is relatively selective. Other reflexes such as pinna and spinal reflexes are only influenced with a dose of two or three times as much.( 30)

The spontaneous motility of experimental animals is lowered only at comparatively high doses. In mouse motility tests the following were ascertained to be the lowest effective doses after i.p. application: 45 mg/kg methysticin and yangonin, 60 mg/kg kavain and dihydromethysticin, 90 mg/kg dihydrokavain and 200 mg/kg desmethoxy-yangonin.( 23, 57)


An anesthetic intensifying effect is a basic property of substances that have a depressive effect on the central nervous system; it is characteristic of neuroleptic drugs (except for tranquilizers) and antihistamines. With the kava pyrones, especially dihydromethysticin, this effect is very distinctive. Through combined administration of dihydromethysticin and hexobarbital, periods of anesthesia lasting up to 27 hours could be achieved, while for the corresponding hexobarbital dose a period of anesthesia of about two hours was observed.( 23) In animal experiments dihydromethysticin is therefore superior to phenothiazines, which are used in modern anesthesia technique. With the help of the EEG, it was shown in greater detail that the pyrones not only increase the duration of anesthesia, but, more importantly, that they intensify and deepen the anesthetic effect.( 58)

The anesthesia intensifying action is valid for injected narcotics such as barbiturates, urethane and gluteth imide, as well as inhaled anesthetics such as ether and laughing gas. The methysticins were the most effective (for more details see references 23, 57). The weakest was piperolide:( 30) a dose of 150 mg/kg i.p. (applied 30 minutes before the beginning of the ether application) only increased the length of the anesthesia to double that of the control group. Methysticin caused a comparable effect with 50 mg/kg, kavain with 100 mg/kg, meprobamate with 75 mg/kg and diazepam with 2 mg/kg.( 23)


In these trials, experimental animals, rats for example, are conditioned to a signal to escape an electrical shock. Under the influence of tranquilizers (and neuroleptic drugs) the flight reaction to the stimulus (signal) is inhibited. The kava extract was tested using a combined avoidance reaction.( 59) Depending on the dosage, inhibitory effects were found for doses over 100 mg extract/kg. No experiments on pure pyrones have been conducted.




The Kava Pyrones and Kava Butenolides as Models for Antiepileptic Drugs

Substances with anticonvulsant properties are potential antiepileptic drugs. But not every substance that acts as an anticonvulsant is usable as an antiepileptic drug. Since these medicines are intended for long-term application, chronic toxic side effects often limit the use Of many anticonvulsants in practical treatment. Currently, as in the past, no antiepileptic drugs intended for long-term treatment exist that have no side effects. With the evidence from several spasm models that kava pyrones have pharmacological effects, an orienting clinical trial was carried out on a kava extract for the first time in 1954.( 60, 61) Unfortunately, with long-term treatment using high doses, an undesired side effect occurs in the form of a yellowing of the skin. As a result, the clinical studies -- what we would today call phase II -- were continued with dihydromethysticin, the kava pyrone with the strongest anticonvulsant effect. Daily doses of 800 mg dihydromethysticin proved to be effective against generalized tonic-clonic seizures (grand mal symptoms), but not against petit mal attacks (absence epilepsy). This long-term treatment using comparatively high doses resulted in a reddening of the conjunctiva after one month of treatment. Additional side effects were frequent vomiting and diarrhea.

In order to widen the therapeutic scope or to "separate" the main effect and the undesired side effects, the dihydromethysticin molecule was chemically varied. It was shown to be important to keep the dioxymethylene group, since the molecule is otherwise too easily metabolized. Even. the loading of methoxy groups led to a decrease in effectiveness. The first main variation consisted of taking away the C( 2) chain between the aryl residue and the pyrone ring (Figure 5). The second change consisted of lengthening the C( 2) chain (Figure 6). The pharmacological screening didn't result in any proof that the molecule variants could be superior to the model dihydromethysticin, which is why no clinical studies have been carried out.( 62) The experiments just described appear to indicate that kava pyrones are not suited for treating epilepsy, which was recently proven again in another context.( 29)

There is, however, a second way to approach the problem of molecule variation. As noted earlier, the piperolide isolated from Mexican Piper sanctum indicates clear differences in the anticonvulsant effect of pyrones:

- anticonvulsant over the entire dosage range (no intensification of tonic cramps in very low doses);

- effect lasts over a long period of time;

- in high doses there is partial alleviation of clonic spasms (reduction of lethality from penetrazol).( 30)

Syntheses were developed for substances of the piperolide and fadyenolide type (e.g. piperolides without a C( 2) chain between the aryl and butenolide rings).( 63-65) If one imagines that the exocyclical double bond of the fadyenolide hydrates and 6-methoxyl saponifies to 6-hydroxyl, one achieves a substance type that exists in two diastereoisomer shapes, as erythry- and threo-derivatives (Figure 7). A diastereoselective representation procedure for manufacturing threo-derivatives was developed in the experimental laboratories of Dr. Willmar Schwabe GmbH & Co., Karlsruhe. The procedure can be carried out using the kilogram-based scale.( 66) Several hundred molecule variants were synthesized and tested pharmacologically for anticonvulsant characteristics. The following connections between chemical structure and anticonvulsant effect were found:( 66)

- the basic requirement is the butenolide structure, which is bound to an aryl residue over a hydroxymethyl group;

- the molecule has to be in the threo form;

- the 4-alkoxy group is essential;

- substitution of the aryl residue in position 2 with electron-attracting substitutes improves the effect;

- additional halogen substitution in position 4 or 5 of the arylrest can lead to a longer effective period.

The substance with the best effective profile is (ñ)-threo-5-( 2-chlorphenyl-hydroxymethyl)-4-methoxy-2(5H)-furanon (Losigamon, Figure 7). The researchers( 66) who discovered this substance selected it for clinical testing (currently phase II to phase III). Further pharmacological and clinical results are cited in references 67-73, 80 and 81.

(1) Lebot V, Lévesque J. The origin and distribution of Kava (Piper methysticum). A phytochemical approach. Allertonia 1989; 5:223-81.

(2) Martinez M. Medical Plants of México (in Spanish). México: Ediciones Botas, 1959:184-5

(3) Manual of Practical and Scientific Pharmacy (in German). Berlin: Urban & Schwarzenberg, 1929; 5:658-61.

(4) Langhammer L, Hänsel R, Goméz-Pompa A. Pharmacognostical characterization of Piper sanctum (in German). Dtsch Apoth Ztg 1972; 112:591-6.

(5) Berger FM, Bradley W. The pharmacological properties of myanesin. Brit J Pharmacol Chemother 1946; 1:265-72.

(6) Snyder SH. Chemistry of the Psyche: Drug Effects in the Brain (in German). 2nd ed. Heidelberg: Spektrum-d. Wiss. Verlagsgesellschaft, 1989: 155-82.

(7) Lewin L. On Piper methysticum (Kava) (in German). Berlin: Verlag August Hirschwald, 1886.

(8) Ching-yao C. A new chemical analysis of narcotic peppers (in German) [disssertation]. Berlin: Freie Universität, 1979.

(9) Borsche W, Blount BK. Experiments on the constituents of the kava root (in German). XIII. Mitteil Chem Ber 1933; 66:803-06.

(10) Hänsel R, Klaproth L. Activated charcoals as adsorbent substances for chromatographical separation of lipophile plant extracts (in German). Arch Pharmaz 1965; 299:71-80.

(11) Hänsel R, Beiersdorff H-U. Notes on the sedative principles of the kava rhizome (in German). Arzneimittelforsch 1959; 9:581-85.

(12) Hänsel R, Klaproth L. Isolation of 11-methoxy-yangonin from kava root (in German). Arch Pharmaz 966; 299:503-06.

(13) Achenbach H, Karl W, Pregel W. Chem Ber 1971; 104:2688. Cited from Ching-yao Chuang (reference 8) (in German):12.

(14) Achenbach H, Karl W, Pregel W. Chem Ber 1972; 105:2182. Cited from Ching-yao Chuang (reference 8) (in German):12.

(15) Lazar J. Article on the chemical composition of kava extract (in German) [dissertation]. Berlin: Freie Universität, 1983.

(16) Hänsel R, Beer C, Schultz J. Two chalkone pigments from Piper methysticum (in German). Zeitschr Naturforsch 1963; 18b:370-73.

(17) Achenbach H, Karl W. Chem Ber 1970; 103:2535. Cited from Ching-yao Chuang (reference 8) (in German):12.

(18) Sauer H, Hänsel R. Kava lactones and flavonoids from a Piper species endemic to Papua, New Guinea (in German). Planta Med 1967; 15:443-58.

(19) Hänsel R, Beer C, Schulz J. A new kava pyrone from Piper sanctum (in German). Chem Ber 1973; 106:3119-26.

(20) Hänsel R, Pelter A. Cinnamylide butenolide from Piper sanctum (in German). Phytochemistry 1971; 10:1627-34.

(21) Kretzchmar R, Teschendorf HJ. Pharmacological research into the sedative-tranquilizing effect of narcotic peppers (in German). Chemiker Ztg 1974; 98:24-28.

(22) Meyer HJ, May HU. Local anesthetic properties of natural kava pyrones (in German). Klin Wochenschr 1964; 42:407-9.

(23) Meyer HJ. Pharmacology of the kava drug (in German) [postdoctoral thesis]. Freiburg: Albert-Ludwig-Universität, 1966.

(24) Mutschler E. Drug effects (in German). Manual of Pharmacology and Toxicology. 5th ed. Stuttgart: Wissenschaftliche Verlagsgesellschaft, 1986: 368.

(25) Eichholtz F, Fleckenstein A, Muschawek R. On the influence of the Bezold-Jarisch reflex by intravenously introduced local anesthetic and antihistamine bodies (in German). Klin Wochenschr 1949; 27:71.

(26) Kretzschmar R and Meyer HJ. Comparative experiments on the anticonvulsant efficacy of Piper methysticum pyrone bonds (in German). Arch Int Pharmacodyn 1969; 177:261-77.

(27) Kretzschmar R, Meyer HJ, Teschendorf HJ, Zöllner B. Antagonistic effect of natural 5,6-hydrated kava pyrones on strychnine poisoning and experimental local tetanus (in German). Arch Int Pharmacodyn 1969; 182:251-68.

(28) Meyer HJ, Kretzschmar R: Research on the relationships between molecular structure and pharmacological effect of aryl-substituted 4-methoxy pyrones of the kava pyrone type (in German). Arzneimittelforschg 1969; 19:617-23.

(29) Kretzschmar R. Pharmacological research on the central-nervous effect and on the effective mechanism of the kava drug (Piper methysticum) and its crystalline constituents (in German). In: Loew D, Rietbrock N, pubs. Phytopharmaceuticals in Research and Clinical Use. Darmstadt, FRG: Steinkopff, 1955:28-38.

(30) Khezerlu Sh. On the pharmacology of lactone bonds. Neuropharmacological study on a cinnamylide butenolide from Piper sanctum (in German) [dissertation]. Freiburg: Albert-Ludwig-Universität, 1971.

(31) Tanaka K. Anticonvulsant properties of procaine, cocaine, adiphenine and related structures. Proc Soc Exp Biol Med 1955; 90:192-5.

(32) Kretzschmar R, Meyer HJ. The influence of natural 5,6 hydrated kava pyrones on isolated heart preparations and their antifibrillatory effect on the entire animal (in German). Arch Int Pharmacodyn 1968; 175:1-15.

(33) Krieglstein J, Oberpichler-Schwent H, pubs. Pharmacology of Cerebral Ischemia. Stuttgart: Medpharm Scientific Publication, 1994.

(34) Backhauá C, Karkoutly C, Welsch M, Krieglstein J. A mouse model of focal cerebral ischemia for screening neuroprotection drug effects. J Pharmacol Meth 1992; 27:27-32.

(35) Fern R, Ransom BR, Waxman SG. Ca(2+)-mediated injury in anoxic white matter: autoprotective mechanisms and therapeutic strategies. In: Krieglstein J, Oberpichler-Schwenk H, pubs: loc. cit. (reference 33): 31-45.

(36) Obrenowitch TP, Urenjak J. Na(+) channel blockade: An underrated approach to cerebroprotection. In: Krieglstein J, Oberpichler-Schwenk H, pubs: loc. cit. (reference 33): 551-63.

(37) Backhauá C, Krieglstein J. Extract of kava and its methysticin constituents protect brain tissues against ischaemic damage in rodents. J Pharmacol 1992; 215:265-9.

(38) Kretzschmar R, Meyer HJ, Teschendorf HJ, Zöllner B. Antispasmodic efficacy of aryl-substituted a-pyrones and aqueous extracts from Piper methysticum (in German). Arch Int Phaarmacodyn 1969; 180:475-91.

(39) Singh YN. Effects of kava on neuromuscular transmission and muscle contractility. J Ethnopharm 1983; 7:267-76.

(40) Maeno T. Analysis of sodium and potassium conductances in the procaine endplate potential. J Physiol (London) 1966; 183:592-606.

(41) Steinbach AB. Alteration by xylocaine (lidocaine) and its derivatives of the time course of the endplate potential. J Gener Physiol 1968; 52:144-61.

(42) Dudel J. Information transfer through electrical stimulus (in German). In: Schmidt RF, Thewes G, pubs. Physiology of Humans. 26th ed. New York: Springer, 1995: 20-42.

(43) Clarenbach P, Fröscher W. Anti-epileptic drugs (in German). In: Riederer P, Laux G, Pöldinger W, pubs. Vienna: Springer Verlag, 1993; 6:57-84.

(44) Ginsberg MD, Sternau LL, Globus MY-T, et al. Therapeutic modulation of brain temperature: relevance to ischaemic brain injury. Cerebrovasc Brain Metab Rev 1992; 4:189-225.

(45) Kass IS, Abramovicz AE, Cottrell JE, Chambers G. The barbiturate thiopental reduces ATP levels during anoxia but improves electrophysiological recovery and ionic homeostasis in the rat hippocampal slice. Neuroscience 1992; 49:537-43.

(46) Gleitz J, Beile A, Peters T. Kavain inhibits Veratrin-activated voltage-dependent Na(+)-channels in synaptosomes prepared from rat cerebral cortex. Neuropharmacology 1995; 34:1133-8.

(47) Bader H, pub. Manual of Pharmacology and Toxicology (in German). 2nd ed. Weinheim: Wissenschaftliche Verlagsgesellschaft, 1985:394,417.

(48) Kretzschmar R, Meyer HJ, Teschendorf HJ. Strychnine antagonistic potency of pyrone compounds of kava root. Experientia 1970; 26:283-4.

(49) Loewe S. Isobols of combined effects of strychnine and barbiturates. Arch int Pharmacodyn 1958; 14:451-64.

(50) Monograph (in German) from the Commission E on "Piperis methystici rhizoma" (kava rootstock) published in the Bundesanzeiger. 101: 6/1/1990.

(51) Volz HP, Hänsel R. Kava-kava and kavain in psychopharmacotherapy. Psychopharmakotherapie 1994; 1:33-9.

(52) Spree MH, Croy H-H. Antares, a standardized kava preparation with the special extract KW 1491 (in German). Der Kassenarzt 1992; 17:44-51.

(53) Siegers CP, Honold E, Krall B, et al. Results of observational studies of L 1090 with Laitan capsules. Ärztliche Forschg 1992; 39:7-11.

(54) Pödlinger W. Compendium of Psychopharmacotherapy. 2nd ed. Granzach/Baden: Wissenschaftlicher Dienst "Roche," 1971: 73.

(55) Kretzschmar R. The significance of the pyrone bonds for the occurence of the sedative effect of narcotic peppers (in German) [Postdoctoral thesis]. Freiburg/Br.: Albert-Ludwig-Universität, 1970.

(56) Kretzschmar R, Teschendorf HJ, Ladous A, Ettahadieh D. On the sedative action of the kava rhizome. Acta Pharmacol Toxicol 1971; 29 Suppl 4:226.

(57) Hölzl J, Juretzek W, Schneider G, Stahl-Biskup E. Piper. In: Hänsel R, Keller K, Rimpler H, Schneider G, pubs.: Hager's Manual of Pharmaceutical Practice (in German). 5th ed. New York: Springer, 1994: 191-220.

(58) Meyer HJ. Pharmacology of the active principles of the kava rhizome. Arch Int Pharmacodyn 1962; 138:505-36.

(59) Duffield PH, Jamieson DD, Duffield AM. Effect of aqueous and lipid-soluble extracts of kava on the conditioned avoidance-response in rats. Arch Int Pharmacodyn 1989; 301:81-90.

(60) Efron DH, ed. Ethnopharmacological Search for Psychoactive Drugs. New York: Raven Press, 1967.

(61) Pfeiffer CC, Murphee HB, Goldstein L. Effect of kava in normal subjects and patients. In: Efron DH, ed.: loc. cit. (reference 60): 155-61.

(62) Klohs MW. Chemistry of kava. In: Efron DH, ed: loc. cit. (60):126-32.

(63) Pelter A, Al-Bayati RIH, Hänsel R, et al. The structure and synthesis of fadyenolide, a new butenolide from Piper fadyenii. Tetrahedron Lett 1981; 22:1545-8.

(64) Pelter A, Ayoub MT, Schulz J, et al. The synthesis of piperolide and related compounds. Tetrahedron Lett 1979; 118:1627-30.

(65) Pelter A, Al-Bayati RIH, Ayoub MT, et al. J Chem Soc Perkin Trans 1987; I:717-42.

(66) Klessing K, Chatterjee SS. á-methoxybutenolide as a potential anticonvulsant: diastereoselective synthesis and structure-effect relationships (in German). Intern. Symposium on "Three-dimensional structure and effect," 1984 October 2-4; Schleiersee (FDR) and patent applications EP-B1 0247320, DE-A 3615157, USUS 4855320, JP-B1 0349609, USUS 4965283.

(67) Chatterjee SS, Klessing K, Stein U. Book of Abstracts. 18th International Epilepsy Congress; 1989; New Delhi, India; 23:6.

(68) Dienel A, Biber A, Klessing K, Stein U. Losigamon: First clinical results. Epilepsia 1992:33 Suppl 3:81.

(69) Dimpfel, W, Chatterjee SS, Nöldner M, Ticku MK. Effects of the anticonvulsant losigamon and its isomers on the GABA-A receptor systom. Epilepsia 1995; 33 Suppl. 3:81.

(70) Köhr G, Heinemann U. Anticonvulsant effects of tetronic acid derivatives on picotoxin induced epileptiform activity in rat hippocampal slices. Neurosci Lett 1989; 112:43-7.

(71) Köhr G, Heinemann U. Effects of tetronic acid derivatives A0-33 (Losigamon) and A0-78 on epileptiform activity and on stimulus-induced calcium concentration changes in rat hippocampal slices. Epilepsy Res 1990; 7:49-58.

(72) Zhang CL, Chatterjee SS, Stein U, Heinemann U. Comparison of the effects of losigamon and its isomers on maximal electroshock-induced epileptiform activity in slices of the rat temporal cortex. Naunyn-Schmiedeberg's Arch Pharmacol 1992; 345:85-92.

(73) Biber A, Dienel A. Pharmacokinetics of losigamon, a new antiepileptic drug, in healthy male volunteers. Int J Clinical Pharmacol Therapeut 1996; 334:6-11.

(74) Hänsel R, Woelk H. Drug Therapy Today: Spectrum Kava-Kava (in German). 2nd ed. Basel: Aesopus-Verlag, 1995.

(75) Faust V. Anxiety states and kava (in German). Apotheker Journal 1995; 17:13-16.

(76) Hänsel R, Kava extract: a natural anxiolytic drug (in German). Apotheker Journal 1993; 15:22-28.

(77) Gerster G, Gracza L. Piper methysticum: data on pharmacology and clinical use of a phytotranquilizer (in German). Therapeutikon 1990; 4:257-66.

(78) Sing YN. Kava: An overview. J Ethnopharmacol 1992; 37:13-45.

(79) Lebot V, Merlin M, Lindstrom L. Kava, The Pacific Drug. New Haven, CT: Yale University Press, 1992.

(80) Yokoyama H, Onodera K, Iinuma K, Watanabe T. Effect of thioperamide, a histamine H(3)-receptor antagonist on electrically induced convulsions in mice. Europ J Pharmacol 1996; 234:129-33.

(81) Schmitz D, Zang CL, Chatterjee SS, Heinemann U. Effects of methysticin on three different models of seizure like events studied in rat hippocampal and entorhinal cortex slices. Naunyn-Schmiedeberg's Arch Pharmacol 1995; 351:348-55.

Natural Product Research Consultants, Inc.


By A. Clay and R. Reichert

Share this with your friends