Hepatotoxicity of the phytomedicines Kava Kava and Cimicifuga Racemosa

In this thesis the hepatotoxic properties of the two phytomedicines kava kava and cimicifuga
racemosa have been investigated. This topic is a prevailing problem as herbal medicines are
quite popular today and not much is known about their toxicological profile. Especially hepatotoxicity
is a wide-spread problem of medicines in general as most of them are metabolized in
the liver and the liver therefore represents a near target. Ingestion of kava has been associated
with liver damage and therefore Kava has been withdrawn from the market in different countries.
For cimicifuga so far not much evidence for liver toxicity in humans exists. However, lately
it was found that rats fed with high doses of cimicifuga developed microvesicular liver steatosis.
In order to examine the in vitro toxicity of the two herbs the hepatocarcinoma cell line
HepG2, the hepatoblastoma cell line HUH6, and isolated rat liver mitochondria have been
used. Three kava extracts (a methanolic and an acetonic root extract; a methanolic leaf extract)
and an ethanolic cimicifuga extract have been investigated.
In the first project (chapter 6) toxicity of kava kava was investigated. It was found that all
three kava extracts showed concentration-dependent toxicity in the general cytotoxicity tests.
Experiments in mitochondria revealed that the kava extracts inhibited and uncoupled (root
extracts) or only uncoupled (leaf extract) the respiratory chain and decreased the mitochondrial
membrane potential; in addition, beta-oxidation was inhibited. Furthermore, oxidized glutathione
was slightly increased and ATP content of the cells was maintained. Apoptosis was
found in HepG2 cells for all three kava extracts. These findings of mitochondria-related toxicity
(affected respiratory chain, decreased mitochondrial membrane potential and increased
reactive oxygen species (ROS) production) might result in the opening of the permeability
transition pore and consequently in the rupture of the outer mitochondrial membrane, thus
possibly leading to the release of cytochrome c and subsequently to apoptosis. These events
might contribute to kava associated hepatotoxicity, especially in predisposed patients having
mitochondrial damages.
In the second project (chapter 7) two methods for cell death determination were applied
and compared in the hepatoma cell lines HepG2 and HUH6 utilizing kava as death-inducing
agent: the annexin V / propidium iodide (PI) stain and the green fluorescent protein (GFP)-
method. The annexin V / PI stain has already been established as a method to detect apoptosis
and necrosis elicited by kava kava in the study mentioned above. The GFP-method had been
described in different cell lines by other groups and is based upon the phenomenon that GFP
decreases its fluorescence when apoptosis and/or necrosis occur. We found that the
annexin V / PI stain and the GFP-method provided similar results. These results on the one
hand confirm that kava indeed induces cell death; on the other hand they show that the GFPmethod
can also be employed in liver cell lines. Therefore, the GFP-method is suited as an
easy, reliable, cost-effective method to screen substances for their hepatotoxic potential.
In a further study (chapter 8) the toxicity profile of the phytomedicine cimicifuga was
assessed. High doses of cimicifuga extract caused microvesicular liver steatosis in rats. Based
on these findings in vitro experiments were performed to further investigate the hepatotoxic
potential of this plant. General cytotoxicity tests revealed concentration-dependent toxicity. In
mitochondria, cimicifuga extract was able to inhibit b-oxidation, to uncouple the respiratory
chain and to reduce the mitochondrial membrane potential. In addition, HepG2 cells underwent
apoptosis when incubated with cimicifuga. It might therefore be assumed that the impairment of
the respiratory chain causes a decrease of the mitochondrial membrane potential, which in turn
could lead to the opening of the permeability transition pore and to apoptosis. Whereas the
inhibition of the b-oxidation may lead to accumulation of fatty acids and subsequently to liver
steatosis. Toxicity was discovered at concentrations higher than the ones expected in humans.
It is therefore conjectured, that hepatotoxicity in humans only occurs under certain conditions.
Finally, it can be concluded that although kava kava and cimicifuga racemosa displayed
clear toxicity in the in vitro situation it has to be further investigated whether these toxicities can
be extrapolated to humans. Factors which have to be taken into account when estimating the
toxic potential for humans include the concentrations used in the in vitro tests in comparison to
portal venous concentrations in humans after intake of the extracts, the metabolism in the body,
predisposing factors for liver toxicity like age, gender and pre-existing liver diseases, and polymedication.
Nevertheless, these cytotoxicity tests represent a valuable tool for evaluating the
toxic potential of herbal products before they reach the market. Of course, other experiments
(animal experiments, for example) would have to be added, if an assessment corresponding to
the rules of the regulatory offices is sought. These in-depth investigations are needed, if botanicals
are to persist on the market in future.