Abbreviations
Abbreviations
ACC: antigen-dependent complement-mediated cytotoxicity
ADCC: antibody-dependent cell mediated cytotoxicity
APN: aminopeptidase N
ARE: antioxidant response element
BAEC: bovine aortic endothelial cell
CAM: chick embryo chorioallantoic membrane
CAT: catalase
COX-2: cyclooxygenase-2
DMBA: 7,12-dimethylbenz(a)anthracene
DMSO: dimethylsulfoxide
ECM: extracellular matrix
FGF: fibroblast growth factor
GM‐CSF: granulocyte-macrophage colony-stimulating factor
GSH: glutathione
HUVEC: human umbilical vein endothelial cell
ID: inhibition dose
IL: interleukin
iNOS: inducible NO-synthase
LD: lethal dose
NADP+: nicotinamide adenine dinucleotide phosphate
NF-κB: nuclear factor-kappaB
NO: nitric oxide
Nrf2: nuclear factor E2-related factor 2
PCNA: proliferating cell nuclear antigen
PGE2: prostaglandin E2
PLA2: phospholipase A2
PPAR: peroxisome proliferator activated receptor
ROS: reactive oxygen species
SOD: superoxide dismutase
Sp: specificity protein
TE: triterpene dry extract
TNF: tumour necrosis factor
TPA: 12-O-tetradecanoylphorbol-13-acetate
TRPC6: transient receptor potential channel 6
VEGF: vascular endothelial growth factor
Introduction
Introduction
Today solid tumours are no longer considered as a mere accumulation of abnormal, malignant
cancer cells. The tumour environment, the tumour stroma, is becoming more and more
important. Therefore, treatment strategies have to be changed. Most therapies now
try to eliminate cancer cells by inducing apoptosis or necrosis. New therapy strategies
include the treatment of the tumour environment, avoiding angiogenesis and modulating
the immune response or the chronic inflammation that is often associated with cancer
promotion and progression. Another approach is to redifferentiate proliferating tumour
cells.
In addition, chemoprevention is important to avoid cancer promotion. Cancer results
from a multistage carcinogenesis process: initiation, promotion and progression. Because
reducing the initiation phase to a zero level is impossible, the most efficient intervention
would be at the promotion phase to eliminate premalignant cells before they become
malignant [1]. Therefore, the concept of delaying or preventing this transformation is worth testing
in future studies [2].
Pentacyclic triterpenes are secondary plant metabolites which arise from cyclization
of squalene [3]. This article focused on triterpenes of the lupeol, oleanane and ursane type. They
are found in different plant organs, e.g., in bark, cork, or in the wax covering leaves
or peel. Low amounts (< 0.1 % of the dry weight of a plant organ) are ubiquitously
present in plants. However, there are a few species that display a high amount of
these pentacyclic triterpenes (> 1 % of the dry weight of the plant organ). The highest
triterpene amount has been found in the outer bark of white birch. The white outer
bark contains up to 34 % (w/w) betulin [4]. Beside the outer bark of birch, leaves of rosemary and olive, mistletoe as well
as plane tree bark and apple peelings contain more than 1 % (w/w) of these pentacyclic
triterpenes ([Table 1]). These plants can be used to obtain triterpene dry extracts consisting of 50–90 %
(w/w) triterpenes [5]. Depending on the plant material, lupeol, betulin, betulinic acid, oleanolic acid,
ursolic acid or an equal mixture of these substances are the main components of such
dry extracts [6] ([Fig. 1]). This kind of triterpene extract may be used as starting material for further pharmaceutical
development.
Fig. 1 Chemical structures of (A) lupeol, (B) betulin, (C) betulinic acid, (D) erythrodiol, (E) oleanolic acid and (F) ursolic acid. Lupeol, betulin and betulinic acid belong to the lupan type pentacyclic
triterpenes consisting of four six-rings and one five-ring, whereas erythrodiol and
oleanolic acid exhibit an oleanan structure with five six-rings. This is quite similar
to the ursan group, which is represented by ursolic acid. The difference between the
oleanan and the ursan type is the methyl-group localization of the E-ring.
Table 1 Plants which display a high amount of pentacyclic triterpenes.
|
Plant
|
Part
|
Triterpene
|
Amount
|
Reference
|
|
Betula alba L., Betulaceae
|
bark
|
lupeol
|
1–2 %
|
[5]
|
|
|
|
betulin
|
10–34 %
|
[4], [5]
|
|
|
|
betulinic acid
|
0.5–1.5 %
|
[5]
|
|
|
|
oleanolic acid
|
0–1.5 %
|
[5]
|
|
Rosmarinus officinalis L., Lamiaceae
|
leaves
|
betulinic acid
|
1.5 %
|
[6]
|
|
|
|
oleanolic acid
|
1.2 %
|
[6]
|
|
|
|
ursolic acid
|
3.0 %
|
[6]
|
|
Malus domestica Mill., Rosaceae
|
fruit peel
|
ursolic acid
|
2.0 %
|
[6]
|
|
Platanus L., Platanaceae
|
bark
|
betulinic acid
|
2.4 %
|
[6]
|
|
Viscum album L., Viscaceae
|
sapling
|
oleanolic acid
|
1.0 %
|
[6]
|
|
Olea europaea L., Oleaceae
|
leaves
|
oleanolic acid
|
3.1 %
|
[6]
|
|
Nerium oleander L., Apocynaceae
|
leaves
|
ursolic acid
|
1.2 %
|
[6]
|
|
Arctostaphylos uva-ursi L., Ericaceae
|
leaves
|
ursolic acid
|
1.2 %
|
[6]
|
|
Coffea L., Rubiaceae
|
leaves
|
ursolic acid
|
1.8 %
|
[6]
|
|
Eucalyptus L'Hér., Myrtaceae
|
leaves
|
ursolic acid
|
1.2 %
|
[6]
|
|
Lavandula angustifolia L., Lamiaceae
|
leaves
|
ursolic acid
|
1.6 %
|
[6]
|
|
Salvia officinalis L., Lamiaceae
|
leaves
|
ursolic acid
|
1.8 %
|
[6]
|
|
Syzygium aromaticum L., Myrtaceae
|
flowers
|
oleanolic acid
|
1.6 %
|
[6]
|
|
Thymus vulgaris L., Lamiaceae
|
leaves
|
ursolic acid
|
1.0 %
|
[6]
|
In the last 15 years hundreds of publications have highlighted the broad spectrum
of biological activities of lupane, oleanane and ursane triterpenes. The literature
search for this review is based on an actual PubMed search focused on the last two
years. On account of the low water solubility of triterpenes, special attention was
given to the concentrations used in in vitro experiments. Concentrations above 100 µM often bias the results, because of an insoluble
fraction. In case of in vivo data, we included effects that are described by almost all different work groups.
Because of their cytotoxicity against various cancer cell lines the group of lupane,
oleanane and ursane triterpenes are considered as promising anticancer drugs. Nevertheless,
due to their various pharmacological activities including antiangiogenic, anti-inflammatory
as well as antioxidant effects and the ability to enhance cell differentiation, they
are more than a simple cytotoxic anticancer drug and are suitable for modern cancer
strategies ([Fig. 2]). Moreover, they are regarded as essential parts of human nutrition because of their
chemopreventive potential to fend off cancer promotion [7], [8].
Fig. 2 Pentacyclic triterpenes such as lupeol, betulin, betulinic acid, erythrodiol, oleanolic
acid and ursolic acid exhibit various pharmacological activities. Besides targeting
tumour cells by induction of apoptosis, they also modulate the tumour environment
displaying antiangiogenic, anti-inflammatory as well as antioxidant effects and enhancing
cell differentiation. The efficacy of each single compound might differ with respect
to the various activities, so the combination of these substances might have a benefit
in order to treat cancer from different from different angles in modern cancer strategies.
Apoptosis
Apoptosis
Induction of apoptosis by pro-apoptotic agents is one important part of cancer therapy.
But apoptosis in cancer cells is often impaired or even blocked by mutated genes regulating
the cell cycle or an imbalanced ratio of pro- and antiapoptotic proteins. Therefore
it is necessary to target different steps of the apoptotic process to bypass such
blocks with respect to the type of cancer. This review highlights only a few aspects
of the knowledge about triterpenes and apoptosis. But it should give an impression
of the diversity of mechanisms triggered by these triterpenes and with it the chance
to overcome apoptosis resistance in cancer cells.
Triterpenes trigger apoptosis by different modes of action, as extensively described
in a series of reviews, especially for betulinic acid [7]–[9]. First it was assumed that betulinic acid is a selective cytotoxic compound against
melanoma cells. However, up to now a large panel of cancer cell lines have proven
to be sensitive to betulinic acid and other pentacyclic triterpenes. It is also assumed
by some authors that there is a selective sensitivity against malignant cells. Nevertheless,
cytotoxicity against primary cells such as fibroblasts, melanocytes, keratinocytes,
neuronal cells and peripheral blood lymphocytes is reported, but they seem to tolerate
higher triterpene concentrations than cancer cells of the same origin [10]–[14]. Whether this may result in a positive effect in vivo, when cancer cells are in a united cell structure is questionable. But in the case
of triterpene acids another possibility to enhance their activity in cancer tissue
was observed. In vitro the activity of betulinic acid was increased by decreasing the pH [15]. And interestingly in athymic mice carrying human melanoma xenografts, its highest
concentration after intraperitoneal injection (500 mg/kg) compared with other tissues
like liver, lung, and kidney was found in the tumour tissue, which often exhibits
a lower pH, caused by a changed metabolism. This could be an explanation of the triterpene
acid accumulation in the melanoma tissue. Up to now, only a few investigations include
pH variations [15], [16], but this fact could be an important factor for increasing the activity of triterpene
acids in cancer treatment and should be investigated in more detail.
The apoptosis mechanism of betulinic acid has been investigated quite well and was
reviewed in 2009 by Fulda [9]. In short, betulinic acid induces apoptosis via the intrinsic pathway by affecting
the mitochondrial membrane potential [17] and initiates reactive oxygen species (ROS) generation linked to an activation of
pro-apoptotic p38 MAPK and SAP/JNK kinases [18], [19]. A similar increase of ROS was also observed for oleanolic acid (25 µM) in astrocytoma
cell lines [20]. While recently published data reported Bax/Bak-independent apoptosis induction
by betulinic acid in various cancer cell lines [21], a number of publications show a modulation of anti- and pro-apoptotic proteins
of the Bcl-2 family [13], [22]–[25]. The modulation of pro- and antiapoptotic factors is complex and probably cell-type-dependent.
It is likely that context dependency also plays a role with respect to nuclear factor
kappa-B (NF-κB) modulation. While NF-κB is activated by betulinic acid (20 µM) in a variety of cancer cell lines resulting
in induction of apoptosis [26], NF-κB inhibition is observed in chemoresistant androgen-refractory prostate cancer cells
exhibiting constitutive Rel/NF-κB activation [27]. Similar effects of triterpenes on NF-κB related to inflammation have been observed and are discussed later. One important
detail to overcome some types of apoptosis resistance is the independence of betulinic
acid induced apoptosis of p53 that is frequently mutated in cancer cells [13], [22], [24].
While apoptosis induced by betulinic acid seems to be independent of the Fas receptor
[22], lupeol targets this receptor and consequently activates the extrinsic pathway via
caspase 8. For example, lupeol (20 µM) significantly increased the expression of the
FADD protein and the Fas receptor in androgen sensitive prostate cancer cells [28]. Furthermore, lupeol sensitises chemoresistant human pancreatic cancer cells (PaC),
to undergo apoptosis by recombinant TRAIL via suppression of cFLIP [29]. Besides, various targets of lupeol are reported to overcome apoptosis resistance
by inhibition of oncogenes and activation of tumour suppressor genes. At a concentration
of 30 µM, lupeol reduces the expression of commonly overexpressed Ras oncoprotein
resulting in the inhibition of the PI3K/Akt pathway that is known for promoting cell
growth [30]. Coincidentally, the expression of phospho-p38 MAPK, which triggers an antiapoptotic
response to tumour cells, was decreased together with NF-κB occurrence. These modulations were accompanied by induction of apoptosis in the
otherwise resistant pancreatic cells [30].
Also ursolic acid and oleanolic acid exhibit pro-apoptotic activity, as reviewed by
Ovesna and colleagues in 2004 [8]. Recent results indicated a modulation of the Bcl-2 protein family due to a suppression
of NF-κB by ursolic acid (50 µM) in B16.F10 mouse melanoma cells. Induction of apoptosis
was accompanied by activation of p53 and caspase-3 gene expression [31]. Oleanolic acid (80 µM) showed apoptosis induction in leukaemia cells (HL60) via
activation of caspase-9 and caspase-3 accompanied by the cleavage of poly(ADP-ribose)
polymerase (PARP) [32].
Betulin has often been found to be inactive or weakly active against several cancer
cell lines such as melanoma (MEL-2), epidermoid carcinoma (KB) [33], leukaemia (HL60, U937, K562) or neuroblastoma (GOTO, NB-1) [34]. However, in nonmalignant, immortalised HaCaT keratinocytes induction of caspase-dependent
apoptosis has been observed [5] and recently, Pyo et al. revealed anticancer activity of betulin (20 µM) against
a human lung cancer cell line (A549) by induction of apoptosis [35]. Erythrodiol, the closely related diol of the oleanane group, has not been investigated
very thoroughly either, but in 2008 apoptotic activity in HT-29 human adenocarcinoma
cells at concentrations of 50–150 µM was reported [36]. In the case of betulin and erythrodiol, it is difficult to evaluate their pro-apoptotic
potential compared to betulinic acid, because of the low amount of published data.
Sometimes only moderate pro-apoptotic effects of triterpenes are observed in vitro, as reported for immortalised HaCaT keratinocytes or human epidermoid carcinoma cells
(A431) treated with a triterpene dry extract from birch bark containing 80 % betulin
and up to 4 % betulinic acid and smaller amounts of lupeol and oleanolic acid [5]. It was only able to induce a twofold higher apoptosis rate in HaCaT keratinocytes.
At first these results seem to have no relevance for therapeutic treatment. Nevertheless
the triterpene extract was successful in vivo treating actinic keratosis [37].
In summary, the apoptotic pathway for betulinic acid is well known. Triggering the
intrinsic pathway via destruction of the mitochondrial membrane potential and including
MAP kinase and PI3K/Akt pathways, seems to be the mode of action. In respect of the
antioxidative activity discussed later, the induction of ROS species in the case of
apoptosis is highly interesting. Potentially the concentration is the critical parameter
causing apoptosis or an antioxidative effect. Oleanolic acid may act in a way similar
to betulinic acid by activating caspase-9.
Unlike triterpene acids, lupeol triggers the extrinsic pathway via the Fas-receptor.
There is still a lack of data for the diols betulin and erythrodiol, thus a prediction
of their mechanism is not yet possible.
Due to the different mechanisms triggered, the use of different triterpenes, also
in a mixture may increase the chance of overcoming the chemoresistence of tumour cells.
At the moment, this is the conclusion drawn only from the results of various researches
carried out independently. Only well organised widespread analysis of a cancer cell
panel treated with different triterpenes under standardised parameters could generate
really comparable data for determining which triterpene or triterpene mixtures exhibit
the best chance of being active against a particular cancer cell. But this could be
an opportunity for tapping the full potential of triterpenes that induce apoptosis.
Antiangiogenic Effects
Antiangiogenic Effects
Angiogenesis is a key process for the outgrowth of cancer cells and their spread into
other tissues. Therefore, suppressing this process is one important pillar of cancer
treatment. There are four key steps in angiogenesis which are potential therapeutic
targets: degradation of extracellular matrix, migration and proliferation of aortic
endothelial cells and the formation of new blood vessels.
The initial results in 1995 of Sohn and colleagues provided an indication that ursolic
acid and oleanolic acid have antiangiogenic effects on bovine aortic endothelial cells
(BAEC) in the CAM (chick embryo chorioallantoic membrane) assay [38]. Here, ursolic acid (ID50: 4 µM) was more effective than oleanolic acid (ID50: 40 µM). However, the key steps of angiogenesis targeted by triterpene acids had
not been identified at this time. Further investigations focussing on the different
key steps disclosed effects on the angiogenic process but these proved controversial.
In serum free cultures of human umbilical vein endothelial cells (HUVECs), ursolic
acid (10–100 µM) increases expression of adhesion molecules that support angiogenesis,
such as ICAM-1 and CD31, and the expression of angiogenic growth factors, particularly
vascular endothelial growth factor (VEGF) and fibroblast growth factor-2 (FGF-2) [39]. This suggests a possible support of the migration step and the structure formation
process. Furthermore, 4–20 µM ursolic acid failed to produce a significant inhibition
of the invasion capability of BAEC through matrigel. In addition, the degradation
of extracellular matrix (ECM) by ECM degradation proteins as MMP-2 and urokinase was
assumed to be stimulated by ursolic acid (4 µM) due to an increased expression shown
by gelatinase and urokinase zymography of these enzymes in BAEC [40]. But this should be investigated in more detail, considering that two years later
Jedinak found a strong inhibition of urokinase activity by ursolic acid in a cell
free system [41]. Despite the enhancement of pro-angiogenic factors, in 2004 Cardenas confirmed Sohn's
(1995) observation of the anti-angiogenic effect of ursolic acid in the CAM assay.
Furthermore, ursolic acid treatment of HUVECs and rat aortic rings that were stimulated
by cultivation in medium supplemented with serum in contrast to serum free medium
surprisingly caused inhibition of the angiogenic phenotype, including the formation
of a capillary network-like structure by HUVECs and a greater extent of endothelial
sprouting in rat aortic rings [39]. Due to the sometimes contradictory effects of ursolic acid on different steps of
angiogenesis, one must be cautious concerning its antiangiogenic potential. The generation
of in vivo data is necessary in order to include the influence of the tissue milieu and thus
appropriately evaluate its effects.
For betulinic acid, older studies revealed that it inhibits enzymatic activity of
aminopeptidase N (APN) in a cell free system [42]. APN is a widely distributed, membrane-bound, zinc-dependent metalloproteinase that
is known to play an important role in tumour-vasculogenesis and is essential for the
endothelial cell tube formation [43], [44]. Betulinic acid (2 µM) potently inhibits basic fibroblast growth factor (bFGF)-induced
invasion and tube formation of BAECs [45]. Initially it was assumed that APN could be the target of betulinic acid. Kwon et
al. confirmed that betulinic acid strongly inhibits enzymatic activity of APN in a
cell free system, but not when enzymatic activity was measured in betulinic acid treated
endothelial cells. He provided evidence that the antiangiogenic activity of betulinic
acid was accompanied by modulation of the mitochondrial membrane function by decreasing
the mitochondrial redox potential. This effect could be blocked by different mitochondrial
permeability transition inhibitors such as cyclosporine A or bongkrekic acid. In view
of the effects of betulinic acid on the mitochondrial membrane, this seems to be a
target structure worth considering. But in this context, the modulation of the mitochondrial
membrane does not cause the release of apoptogenic factors that directly trigger cell
death. Betulinic acid at up to 9 µM did not affect the endothelial cell viability
in the formed tubes. Kwon hypothesised that betulinic acid had a specific effect on
the angiogenic differentiation of endothelial cells, rather than an antiproliferative
activity [45]. It is known, that a modulation of the mitochondrial oxidative phosphorylation can
enhance angiogenic differentiation of endothelial cells, stopping their proliferative
activity [46]. However, in a human prostate cancer cell line (LNCaP) and in vivo, betulinic acid acts via decreasing expression of VEGF [47]. Thus there might be also an antiproliferative effect. The mechanism might be due
to activation of selective proteasome-dependent degradation of the transcription factors
specificity protein 1 (Sp1), Sp3, and Sp4 which regulate the VEGF expression and that
are mostly overexpressed in tumours, as was shown by Chintharlapalli and colleagues
[47]. The concentration-dependent effect on the transcription factors and the expression
of VEGF by betulinic acid could be fully blocked by using a proteasome inhibitor.
Only one publication reports an antiangiogenic effect of lupeol. You and colleagues
found that lupeol also inhibits HUVEC tube formation [48].
Based on the current literature, primarily triterpene acids seem to have an antiangiogenic
effect. However, to clarify the exact mechanisms by which they exert this effect more
experimental work is still needed, even if there are some doubts in the case of ursolic
acid because of the upregulation of pro-angiogenic factors such as MMP-2 or VEGF in vitro. It is necessary to interpret these pro-angiogenic data very carefully. There are
four key steps in angiogenesis. These include degradation of the extracellular matrix,
migration and proliferation of aortic endothelial cells and the formation of new blood
vessels. All four steps are necessary for successful angiogenesis. Most in vitro experiments focus only on parts of the process. This does not enable a prediction
of full angiogenesis. Instead, in vivo models, such as the CAM assay, consider the whole process. Thus they provide information
about the efficacy of the substances on the end result, namely forming new blood vessels,
i.e., not just on one essential factor in a complex network, such as VEGF.
Nevertheless, the in vitro data give hints for understanding the mechanism. In respect of the differentiation
inducing activity of triterpenes discussed later, the most interesting result is that
besides the regulation of endothelial cell proliferation by modulation of growth factors
such as VEGF, the aspect of inducing differentiation to stop proliferation may also
play a role in the angiogenic efficacy of triterpenes.
Anti-inflammatory Effects
Anti-inflammatory Effects
Recent studies have revealed a clear role for inflammation in the development and
progression of cancer and in the immune response against it by orchestrating the tumour
supporting environment [49].
Lupanes, oleananes and ursanes applied orally or topically exhibit significant anti-inflammatory
activity in vivo. This was demonstrated in 12-O-tetradecanoylphorbol-13-acetate (TPA), carrageenan,
serotonin or croton oil induced paw/ear oedema tests, as well as in arthritic animal
models [50]–[56]. Efforts to work out the underlying mechanism in vitro are in progress (reviewed in [7], [8], [57]–[59]) and several potential targets have been discovered.
Besides direct effects on the morphology or the activity of immune cells, such as
macrophages, dendritic cells, T cells or other leukocytes, which may suppress the
immune response [60]–[63], an influence on pro-inflammatory cytokines, e.g., TNF-α, INFγ, IL-1β, IL-6, IL-2, IL-4, IL-5, IL-8, or IL-13 [31], [50], [60], [64]–[66] has been reported.
The expression of these cytokines is regulated by the transcription factor NF-κB, which is therefore a pivotal target. Furthermore, NF-κB is commonly overexpressed in cancer cells. On the one hand this may support the
maintenance of a chronically inflamed microenvironment and on the other hand it often
suppresses apoptosis of the tumour cells [67]. In the last few years, several groups have published controversial data concerning
the activity of betulinic acid and ursolic acid on NF-κB [26], [68]. Kasperczyk et al. [26] postulated an NF- κB activating effect of betulinic acid (13–22 µM) on various cancer cell lines (neuroblastoma,
melanoma, glioblastoma). In contrast, in 2003 Takada and Aggarwal described an inhibition
of NF-κB regulated cyclooxygenase-2 (COX-2) expression and determined a maximal suppressive
effect of betulinic acid at a concentration of 30 µM on NF-κB in colon carcinoma cells [68]. Similarly NF-κB in melanoma cells was inhibited by ursolic acid (50 µM) accompanied by downregulation
of pro-inflammatory cytokines such as TNF-α; IL-1β, IL-6, and GM‐CSF and apoptosis occurred after 48 h [31]. Also, carcinogen-induced NF-κB expression is decreased by ursolic acid [69]. However, in contrast to this, in resting macrophages ursolic acid and also oleanolic
acid activate NF-κB causing increased expression of pro-inflammatory mediators such as TNF-α at concentrations of 5 µM and 4 µM, respectively [70], [71].
At first these results seem to be contradictory, but the different observations could
be based on concentration-dependent effects, as observed for oleanolic acid with respect
to the TNF-α production of human mononuclear cells [65] or for betulinic acid concerning TNF-α and IL-1β production in non-stimulated RAW264.7 macrophages [63]. Another conceivable reason is the influence of the milieu or the cell status that
may crucially modulate triterpene effects. Their investigation on TNF-α or nitric oxide (NO) production via inducible NO-synthase (iNOS) indicate that [63], [71], [72]. Using a stimulated cell system (e.g,. activated macrophages) it was possible to
observe an inhibition of pro-inflammatory mediators by triterpene treatment [63], [72]. However, treatment of non-stimulated cells, such as resting macrophages, with triterpenes
led to an increase of pro-inflammatory factors such as TNF-α or IL-1β [64], [71]. Certainly, these are in vitro data and up to now it is not clear whether these findings are meaningful for in vivo models or in therapeutic use. But triterpene-induced effects seem to be critically
affected by environmental conditions.
Phospholipase A2 (PLA2) provides substrate for cyclooxygenase and 5-lipoxygenase. These pathways are major
pathways of the inflammation process. Betulin and betulinic acid [73], as well as oleanolic acid [74], can inhibit PLA2. Downstream, COX-2 and its product prostaglandin E2 (PGE2) are also repressed by lupeol, betulin, betulinic, and ursolic acid [61], [63], [68], [75], [76]. Again COX-2 expression is regulated by NF-κB, suggesting an inhibitory effect of triterpenes on this transcription factor [68], [77].
Unfortunately, the use of different cell systems, with a diverse metabolic background,
plus the usage of different triterpenes, in various concentrations, precludes suggesting
an exact mode of action for the anti-inflammatory effect of these substances. However,
the abundance of data, and especially the in vivo observations, evidenced the anti-inflammatory potential of the listed pentacyclic
triterpenes. A promising target for the triterpenes presented seems to be NF-κB. A number of proteins modulated by triterpenes, such as TNF-α, IL-8 or COX-2 are under control of this transcription factor. However, the milieu,
such as the cell status, has to be considered because it seems to have a strong influence
on the triterpene effects and should receive special attention.
Antioxidative Effects
Antioxidative Effects
ROS are well recognised as playing a dual role as both deleterious and beneficial
species. ROS are normally generated by tightly regulated enzymes to maintain moderate
concentrations, providing beneficial physiological effects, for example in cellular
responses to noxia, or in the regulation of immune responses. Overproduction results
in oxidative stress that can be an important mediator of damage to cell structures
[78]. Initially increased levels of ROS disrupt cell membrane integrity by oxidation
of unsaturated membrane lipids. Lipid peroxidation is commonly related to cardiovascular
diseases [79], autoimmune diseases or chronic inflammation [80]. Furthermore, free radicals cause DNA damage which may result in tumour initiation
and promotion [81]. Thus regulation of the ROS level may be an important preventive measure and may
also support the anticancer therapies, by avoiding oxidative stress.
The organism uses two antioxidative mechanisms to regulate the level of free radicals,
first an enzymatic and second a non-enzymatic system [82], [83]. The enzymatic system concerns enzymes such as superoxide dismutase (SOD) or catalase
(CAT) that are oxidised and reduced within a cascade to eliminate the free radicals.
The non-enzymatic system deals with antioxidants. One of the body's own antioxidants
is glutathione (GSH). It exists as a monomeric tripeptide (GSH); when oxidised a GSSG
dimer is generated. In order to use the reducing power of glutathione to catalyze
disulfide reductions in the presence of NADPH and glutathione reductase, enzymes such
as glutathione-S-transferase and glutathione peroxidase are necessary [84].
Triterpenes, particularly lupeol, but also betulin and ursolic acid are known for
their antioxidative potential [85]–[90]. They do not act as a classical antioxidant; however, triterpenes activate the enzymatic
system by increasing the activity of SOD as well as CAT and glutathione S-transferase
and glutathione peroxidase [90]–[92].
In detail, lupeol especially displays convincing effects particularly on chronic inflammatory
diseases such as chronic arthritis, but also as a chemoprotective agent. When arthritic
rats were treated orally with lupeol (50 mg/kg body weight daily for 8 days), a significant
decrease of the inflammatory symptoms was observed, while the activity of the antioxidative
enzymes SOD and CAT were elevated [93]. Another positive effect of lupeol is documented in the case of hyperoxaluria in
rats [94]. The excess of oxalate causes a high oxidative stress on the renal tissue. Similar
to lupeol, betulin (35 mg/kg body weight daily for 21 days) normalises the glutathione
status, increases the SOD and CAT activity [90] and decreases the peroxidation of erythrocyte membrane lipids as well as normalises
the activity of membrane bound ATPases [92]. A third indication that has been investigated is hypercholesterolemia. Lupeol (50 mg/kg
body weight daily) normalises the lipid profile and activates the body's own antioxidative
system followed by a decrease of oxidative stress in rats. This results in a protection
of renal tissue. In this case the antioxidative effect of lupeol is called cardioprotective
or renalprotective [87], [95], [96]. In the majority of in vivo studies lupeol and also betulin were administered orally daily without any adverse
effects at a dosage of 35–50 mg/kg body weight, which is a high dose. Considering
an average human body weight of 60 kg, the application of 3 g of the drug per day
would be necessary.
The diseases mentioned are not directly related to the development of cancer, but
they illustrate the antioxidative activity of lupeol or betulin. This seems to be
not only a central part of their biological activity but also the basic mechanism
of their chemopreventive effects which may avoid cancer development.
Substances that cause cell damage, particularly DNA damage or induce chronic inflammations
are potential carcinogens. Agents that shield the organism from these attacks are
called chemopreventive. Recently lupeol was termed as a chemopreventive agent, reviewed
by Chaturvedi and colleagues [7]. This includes hepatoprotective effects protecting liver cells from cadmium, 7,12-dimethylbenz(a)anthracene
(DMBA) or hepatotoxic aflatoxins [97], [98], [99], [100] or cardioprotective activity shielding cardiac tissue from cyclophosphamide induced
cardiotoxicity [88] by oral administration of lupeol. Furthermore, a chemoprotective effect of topically
applied lupeol was observed when skin was treated with benzoylperoxides [89], [101] or DMBA [102]. All these potential carcinogenic agents cause oxidative stress, deplete glutathione
and decrease the activity of antioxidant enzymes. Lupeol regenerates the glutathione
pool along with an elevation in the activities of the antioxidising enzymes and anti-oxidants.
Further cytoprotective effects are also known for betulin and triterpene acids. In vitro pretreatment with 2 and 22 µM ursolic acid protects human lymphocytes against UVB-induced
lipid peroxidation and DNA-damage concentration-dependently [86]. Various in vitro studies revealed hepatoprotective effects of betulin, betulinic, ursolic and oleanolic
acid against cadmium or ethanol-induced toxicity in HepG2 cells (2–11 µM) [103], [104]. It should be mentioned that these effects occur using subtoxic triterpene concentrations
depending on the cell type. Recently published data showed chemopreventive activity
of oleanolic and ursolic acid also in vivo. Rats treated orally with 1,2-dimethylhydrazine developed colon associated carcinogenic
dysplasia caused by agent-induced oxidative stress. Simultaneous oral administration
of 25 mg/kg body weight oleanolic or ursolic acid decreases the appearance of cell
damage [105].
Further experiments focussed on hepatoprotective effects have identified a possible
key target for the antioxidative effect of triterpenes. Oleanolic acid treatment dramatically
increased expression of the transcription factor nuclear factor E2-related factor
2 (Nrf2) [106]. Regulatory regions of the genes for cytoprotective enzymes such as glutathione
S-transferase or SOD contain the antioxidant response element (ARE), which is activated
upon binding of Nrf2. Nrf2 has been shown to be essential in the upregulation of these
genes in response to oxidative stress [107]. The mechanism of increasing Nrf2 expression by oleanolic acid has not been clarified
so far. One hypothesis might be the generation of ROS that triggers the antioxidative
cascade including Nrf2 expression. For oleanolic acid treatment at concentrations
of 25 µM, accumulation of ROS in an astrocytoma cell line resulting in apoptosis has
been reported [20], but it was interpreted only in the context of apoptosis induction. Also betulinic
acid (20–100 µM) showed ROS generation at concentrations resulting in apoptosis [18], [24]. Therefore the question arises, whether generation of ROS could be induced by triterpenes
or especially by triterpene acids used in a subtoxic concentration, and if this could
lead to an activation of the antioxidative system.
Taken together, lupeol and betulin, as well as triterpene acids, such as oleanolic
acid and ursolic acid, display an antioxidative activity. The underlying mechanism
is the modulation of the body's own enzymatic antioxidative system including enzymes
such as SOD and CAT, as well as glutathione S-transferase, perhaps triggered by the
activation of Nrf2 and leading to a generally elevated antioxidant status of the organism.
The antioxidative effect of triterpenes seems to be a central part of their biological
activity and may also be useful as a preventive strategy in the case of cancer.
Redifferentiation
Redifferentiation
Differentiation and proliferation of a cell are mutually exclusive. Differentiation
takes place in the G0 state of a cell. In order to proliferate, reentering the cell
cycle is necessary. Thus, achieving differentiation of a still proliferating cancer
cell is one possible approach of cancer treatment. In this respect, compounds that
influence and enhance differentiation processes in cells are promising candidates
for cancer therapy. In case of triterpenes, very interesting data reveal the promotion
of differentiation of healthy cells such as keratinocytes and also of the redifferentiation
of tumour cells.
With regard to differentiation of healthy cells, it was shown that betulinic acid
induces differentiation in normal keratinocytes in vitro at concentrations of 9–18 µM, including an upregulation of filaggrin and involucrin
[11]. Likewise, oleanolic acid, but not ursolic acid, increases the expression of these
two differentiation markers in vivo in mouse skin disrupted by tape stripping. Oleanolic acid not only upregulated structural
proteins localised in the spinous/granular layers in the epidermis, but also the cornified
envelope formation as the final product of terminal differentiation was increased
and the barrier function of the epidermis was regenerated, as measured by a decreased
transepidermal water loss [108]. Similar results were obtained with a highly purified triterpene dry extract from
birch bark consisting of betulin (80 %), betulinic acid, oleanolic acid, and lupeol
(1–4 %). Normal human keratinocytes and human skin explants were treated with 10 µg/mL
of the extract. Besides the upregulation of early and late differentiation markers
such as loricrin, filaggrin, involucrin, transglutaminase, and keratin 10, as well
as increased Notch2 expression, also a typical degradation of DNA strictly limited
to the distal stratum granulosum cells was observed [109].
One possible signalling cascade activated by triterpenes is the peroxisome proliferator
activated receptor-α (PPAR-α) pathway, which is involved in the regulation of lipid metabolism in the epidermis
[110]. Oleanolic acid and ursolic acid topically applied on disrupted mouse skin improved
the recovery of permeability barrier functions and stimulated epidermal differentiation
via PPAR-α expression [108], [111]. Another hypothesis is the selective upregulation of the transient receptor potential
channel 6 (TRPC6) by triterpenes in primary keratinocytes [109]. TRPC6 is a calcium channel mediating differentiation by regulating the Ca2+ influx into the keratinocytes [112].
Besides the upregulation of differentiation markers, or the activation of corresponding
signalling pathways, morphological changes are also an important part of the differentiation
of many cell types and, therefore, a useful additional marker to recognise differentiation.
Thus, the morphological change of primary keratinocytes into a typical hexagonal corneocyte
shape under triterpene acid treatment encourages the suggestion of their differentiation-inducing
efficacy [11], [108]. Also mammary epithelial cells of rats as well as tumour cells, such as tetracarcinoma
stem cells [113], showed morphological changes when treated with oleanolic acid or ursolic acid [114]. Not only triterpene acids induce such morphological changes, but also treatment
with lupeol results in modifications of the cell shape, e.g., of the mouse melanoma
cell line B16 2F2, and an inhibition of the ability of the cells to migrate, thus
reducing their malignant potential [115]. The morphological changes mediated by alterations of the cytoskeleton are accompanied
by the degradation and reconstruction of microfilaments consisting of actin, whereas
the actin content of the cells remains constant [20], [115], [116]. Up to now it is not clear whether these changes are directly induced by triterpenoids
or are a result of the activated differentiation process. But it should be mentioned
that morphological changes could also be observed in the apoptosis process induced
by triterpenes [20]. Therefore the interpretation of this phenomenon must be carefully evaluated.
Moreover, differentiation is also indicated by regaining or developing special cell
type specific properties. For example, myeloid leukaemia cells (HL-60) amplify 1-alpha,25-dihydroxyvitamin
D3 induced monocytic marker expression such as CD11b or CD14 when treated additionally
with betulinic acid [117]. Oleanolic acid decreases the proliferation rate of M1 mouse carcinoma cells and
human leukaemia cells (HL-60) while phagocytotic activity is increased. And lupeol
as well as betulin and betulinic acid induce melanogenesis in B16 2F2 mouse melanoma
cells [118].
The ability to induce differentiation or redifferentiation, instead of acting as a
cytotoxic substance, is always a question of the available triterpene concentration
and depends also on cell type and cell status [109]. Until now some mechanistic investigations have been carried out only in keratinocytes,
revealing two possible targets: PPAR-α and TRPC6. For other cell types, the observations were only of the differentiation
process or the outcome, for example, morphological changes, developing special abilities
such as phagocytotic activity or expressing differentiation markers such as CD11b
or CD14 in the case of myeloid leukaemia cells. The differentiation data are still
fragmentary, but here the potential of triterpenes seems to be worth investigating
in detail.
Anticancer Activity in vivo
Anticancer Activity in vivo
The relevance of in vitro results can only be judged against subsequent in vivo studies. Because of this, the strong interest on pentacyclic triterpenes as anticancer
agents did not start until betulinic acid was found to be effective in vivo against melanoma by Pisha et al. in 1995 [119].
Bioavailability is the precondition of in vivo effects. Only a few pharmacokinetic studies have been published. BA was found in
various tissues 24 h after i. p. administration (500 mg/kg; mouse) and reached its
highest concentration in perirenal fat. Peak serum concentration of 4.0 µg/mL was
observed at 0.23 h after application [120]. BE reaches a saturation concentration of 138 ng/mL within 4 h after i. p. administration
to rats [121]. These relatively low serum levels can be explained by the low solubility in water
(BA, OA: 0.02 µg/mL and BE < 0.1 µg/mL) [62], [122]. It is known that OA is able to bind to plasmin and albumin [17], [21], so binding phenomena could support the bioavailability. Thus the prediction of
in vivo effects and the concentrations necessary for them may be unreliable if based on in vitro results.
For instance, cytotoxicity was shown for lupeol combating testosterone-induced prostate
enlargement in mice, by inducing apoptosis in the hypodiploid regions, and in tumours
with human prostate origin in a xenograft model [28], [123]. Recent findings showed that lupeol (40 mg/kg body weight thrice a week) also inhibits
the growth of highly aggressive human metastatic melanoma cells (451Lu) in an athymic
nude mouse xenograft model. Immunohistochemical analysis of tumour tissue revealed
that animals receiving lupeol exhibit decreased Ki67 and PCNA-positive cells, suggesting
an antiproliferative effect of lupeol. This correlated with a decreased number of
cyclin D1, cyclin D2 and Cdk2 positive cells and an elevated level of p21 protein
compared to the control mice. The latter result indicates that lupeol causes growth
arrest in melanoma tumours by interfering with the cyclin/cdk2/p21 complex activity
[124].
Modern cancer treatment also includes modulation of the immune system. As noted above,
in vitro a broad spectrum of immune modulations by triterpenes is observed. Indeed, this could
be confirmed in some in vivo studies, e.g., in a study using melanoma (B16.F10) bearing mice (C57BL/6). I. p.
injected ursolic or oleanolic acid (50 µmol/kg body weight, for 5 days) was found
to produce enhanced natural killer cell activity and increased the cytokine IL-2 that
promotes the lytic activity of NK cells. In addition, antibody-dependent cell mediated
cytotoxicity (ADCC) as well as antibody-dependent complement-mediated cytotoxicity
(ACC) were enhanced. According to the expected anticancer effect, the elevated levels
of GM‐CSF and IL-6 in tumour-bearing control animals were also reduced by the treatment
with ursolic acid [125]. Another study showed lupeol efficacy in a TPA induced mouse skin tumourigenesis
model (CD1) by its anti-inflammatory activity. Prior topical application of 1–2 mg/animal
lupeol resulted in the inhibition of the TPA induced activation of PI3K and NF-κB and in an inhibition of COX-2 and iNOS protein expression. The mice showed significant
reduced tumour incidence, lower tumour burden and a delay in the latency period for
tumour appearance [76].
While the antiangiogenic effect of ursolic acid in vitro was still being discussed, Lee et al. (2001) proposed this kind of anticancer effect
based on in vivo studies. Reduced oxygen consumption after treatment as well as a significant decreased
tumour interstitial fluid and blood pressure were obtained after i. p. application
of 100 mg/kg body weight ursolic acid. This was accompanied by an inhibited tumour
growth of a murine fibrosarcoma (FSaII) [126]. The inhibitory action of ursolic acid on urokinase as observed in vitro was assumed to be important with respect not only to the antiangiogenic effect but
also to the suppression of tumour-invasion and metastasis. An in vivo study on B16 mouse melanoma treated C57BL/6 mice showed the complete inhibition of
lung colonisation after 50 mg/kg ursolic acid administered i. p. daily, immediately
after tumour injection during 16 consecutive days. The authors proposed that besides
the in vitro observed urokinase inhibition activity of ursolic acid, an inhibition of cathepsin
B, which represents another possible drug target for the suppression of tumour invasion
and metastasis, may also play a crucial role [41]. Comparable results with respect to tumour metastasis were found for betulinic acid
(10 mg/kg body weight per day) when used in a similar experimental setup [127]. As already mentioned above, betulinic acid targets VEGF expression in a prostate
cancer mouse model and induces the selective proteasome-mediated degradation of transcription
factor specificity protein Sp1, Sp3 and Sp4 [47].
Another approach for the usage of triterpenes in cancer-therapy is their combination
with established chemotherapeutics. In vivo studies using an orthotopic metastatic nude mouse model of oral tongue squamous cell
carcinoma showed that lupeol (2 mg/kg body weight) dramatically decreased the tumour
volume and suppressed local metastasis without side effects. It was 3-fold more effective
than cisplatin, a commonly used chemotherapeutic agent with severe side effects. Surprisingly,
lupeol in combination with low-dose cisplatin was 13-fold more potent than lupeol
alone and up to 40-fold more than cisplatin alone, certainly without side effects
in the animal model used [128]. In another study betulinic acid augmented the inhibitory effect of vincristine,
the major chemotherapeutic agent used for the treatment of melanoma [127]. Combination with triterpenes allowed reduction of the concentration of the chemotherapeutic
agent, without loosing the effectiveness of treatment, while side effects were decreased.
As mentioned previously, up to now there has been only very limited experience in
the treatment of human cancer patients with the listed triterpenoids. It should be
mentioned, that in animal studies dosages of between 10 and 100 mg/kg body weight
are applied which is between 0.6 and 6 g per patient per application (60 kg body weight
assumed). This is an unusually high amount for a drug, and the feasibility and the
relevance of this dose is questionable. Nevertheless, betulinic acid is currently
undergoing a phase II clinical trial for dysplastic melanocytic naevus (web site:
ClinicalTrials.gov). Beside this, recent data show the successful topical treatment
of precancerous lesions, namely actinic keratoses, with a triterpene dry extract (TE)
of the outer bark of birch as mentioned above [37]. Immunohistochemical investigations of biopsies before and after a three month treatment
with TE showed a downstaging of the actinic keratosis and a reorganised epidermal
structure.
It has been possible to confirm in vivo all the general effects of the triterpenes discussed that were predicted based on
in vitro data (inducing apoptosis, anti-inflammatory, antiangiogenic and antioxidative effects),
except for the redifferentiation effect. In this case, the observation of the downstaging
of actinic keratosis lesions after a three month treatment with triterpenes in a clinical
trial has been the only observation available until now, but there is no experimental
work. Thus it is too soon to claim a re-differentiation effect for the triterpenes
in cancer treatment; however it is a promising field of triterpene research.
Administration of drugs also raises the question of their toxicity in vivo. Pentacyclic triterpenes of the lupane, oleanane and ursane group are considered
as relatively nontoxic drugs. A recently published subchronic toxicity study showed
that intraperitoneal and subcutaneous administration of a triterpene mixture (80 %
betulin, betulinic acid, lupeol, oleanolic acid, erythrodiol 1–4 %) produces no toxic
effects [121]. This is in line with previously published data for single triterpenes. For example,
i. p. administered oleanolic acid has a LD50 of 1500 mg/mL in mice [129] and a single s. c. dose of 1000 mg/mL caused no toxic effects in rats [130]. Also i. p. administration of 500 mg/kg body weight betulinic acid in suspension
caused no toxicity [131]. As summarised previously, administration of betulin or lupeol at concentrations
of 35–50 mg/kg body weight also produces no toxic effects.
Conclusion
Conclusion
Cancer is a disease with multiple etiological factors and multiple oncogenes are involved
in its pathogenesis. Therefore, beside a combination of different treatment strategies,
multifunctional agents with multiple targets also offer a more rational approach than
single ones to both its prevention and therapy.
The literature survey reveals that lupeol, betulin, betulinic acid, oleanolic and
ursolic acid are multitarget agents ([Fig. 2]). They fit to the concept of modern cancer therapy, by treating cancer from different
sides, including the tumour environment and the immune system.
But parallel testing of these compounds revealed differences in their efficacy in
several assays. Therefore, the combination of different triterpenes may be a way to
improve their potential as multitarget drugs. Plants like white birch, rosemary, or
mistletoe offer different natural compositions of triterpenes, and their triterpene
extracts may be used as starting material for further pharmaceutical development.
Another possible application is combining triterpenes with already commonly used chemotherapeutic
agents. This may allow lowering the chemotherapeutic dose without loss of efficacy
but hopefully brings with it less adverse effects and may even give synergistic effects.
The pharmacological potential of triterpenes for cancer treatment seems to be high,
although up to now no clinical trial has been published using triterpenes in cancer
therapy. This may be explained by their almost complete insolubility in water [5], [122]. But this galenic problem can be solved by derivatisation, complexation [132], [133], or liposomal formulation [134]. Another problem might be that triterpenes often provide only moderate effects in vitro, perhaps due to their poor solubility [122] and the use of solvents such as dimethylsulfoxide (DMSO) that are not inert. In
our experience, using 1 % DMSO in cell culture medium, a maximum concentration between
20 and 40 µg/ml triterpenes is possible without crystallisation but it has to be carefully
checked in each case. Nonetheless they exhibit convincing effects when applied in vivo, as seen in the case of the birch bark TE extract. Further, the high dose that is
used in animal tests might be an obstacle to transferring it to a therapeutic dose
used in human treatment. Up to now, only two clinical trials have been carried out
with these substances, one treating actinic keratosis and another dysplastic melanocytic
naevus. In each case topical ointments were used. Internal application, such as s. c.
or i. v., has not yet been tested in human cancer treatment.
Looking into future, there remains some tasks to do in order to tap the full potential
of these triterpenes. A lot of effects are not fully understood like the antioxidative
efficacy or the differentiation effects. Another point is that the strength of each
single compound is not defined detailed enough, to find their optimal domain or to
compose them. The differences between the experimental settings, the used concentrations,
time courses or the various cell lines, with different metabolic background and varying
cell culture parameters, makes it difficult to compare and combine results to get
an idea of the underlying mechanisms. Therefore, parallel-testing of triterpenes in
a standardised setting would be preferable. Due to their insolubility in water the
bioavailability is not given in an optimal way, nevertheless, in vivo effects are observed. This may be improved by optimising the galenic form or the
application route.
Their chemopreventive activity makes triterpenes interesting in another field. As
secondary plant metabolites they are present in food. For example, due to the common
use of olives and herbs such as rosemary of the Lamiaceae family, the Mediterranean
diet is high in triterpenes. It is not proven yet, whether they are responsible for
the beneficial effects of this nutrition on health.
In conclusion pentacyclic triterpenes are a rich natural pool of promising anticancer
drugs as well as chemopreventive agents. Further, they are available in high amounts
also in industrial waste products like birch bark or the pulpy residue from apples.
This and their promising pharmacological effects warrants further pharmaceutical development
and clinical investigations to conclude the puzzle of their biological activities.
Acknowledgements
Acknowledgements
The author thanks Stefan F. Martin, Sebastian Jäger, Irmgard Merfort and David J.
Heaf for expert assistance and critical reading in preparation of the manuscript.
