Você está na página 1de 12

Indian Journal of Experimental Biology

Vol 48, March 2010, pp. 208-219

Review Article

Drug development from natural products: Exploiting synergistic effects


Gudrun Ulrich-Merzenicha*, D Panekb, H Zeitlerb, H Vettera& H Wagnerc
a
Medical Policlinic of the Rheinische Friedrich-Wilhelms-University of Bonn, Wilhelmstr. 35-37, D-53111 Bonn, Germany
b
Internal Medical Clinic I (CETA) of the Rheinische Friedrich-Wilhelms-University of Bonn,
Sigmund-Freud-Str. 25, D-53227 Bonn, Germany
c
Department of Pharmacy, Centre of Pharma Research, Ludwig-Maximilians-University,
Butenandstr. 5-13, House B, D-81377 Munich, Germany

Drug development in phytomedicine has been focused in the past on the discovery and analysis of new structures from
natural products. The search aimed at the determination of the single “active principle” in plants, based on the assumption that a
plant has one or a few ingredients which determine its therapeutic effects. But traditional systems of medicines like Ayurveda,
traditional Chinese medicine or the European phytotherapy generally assume that a synergy of all ingredients of the plants will
bring about the maximum of therapeutic efficacy. This approach has for long been impossible to investigate since adequate
methods to standardize complex plant mixtures as well as to rationalize complex mode of actions were lacking. The
introduction of high throughput technologies provides the opportunity to determine profiles of plants and to systematically
explore the mode of action of combinatory drug regimes. The present review highlights the concept of synergy and gives
examples of synergistic effects of plant constituents. It elaborates on how the high throughput technologies can be used in drug
development from natural products with the aim of creating evidence-based plant medications in prevention and treatment of
different diseases in the form of new single treatments or new combinatory drug regimes while exploiting synergy-effects.

Keywords: High-throughput technologies, Phytomedicine, Synergy

Phytobotanical and ethnobotanical research have regained an increasing interest in drugs from
focused for decades mostly on the search for the natural resources for different reasons:
single “active principle” in plants, based on the
assumption that a plant has one or a few ingredients a) Large screening projects for the isolation
which determine its therapeutic effects. But and characterization of single bioactive
traditional systems of medicines like Ayurveda, compounds yielded only moderate results, e. g.
traditional Chinese Medicine (TCM) or the The National Cancer Institute of the United
European phytotherapy generally assume that a States screened about 114,000 extracts from an
synergy of all ingredients of the plants will bring estimated 35,000 plant samples against tumor
about the maximum of therapeutic efficacy. This systems4. Clinically significant cancer
approach has for long been impossible to chemotherapeutic agents included Paclitaxel
investigate since adequate methods to standardize (tamoxifen), topotecan and CPT-114. A
complex plant mixtures (as drugs) as well as to provision for synergistic effects might elevate
rationalize the complex mode of actions were the yield.
lacking. The recent development in synergy b) The major burden to any health system is the
research and the omic-technologies have opened treatment of multifactorial diseases with
highly interesting perspectives for a new generation heterogenous disease courses (e.g. cardiovascular
of phytopharmaceuticals. The present review is diseases, diabetis type 2). The efficiacy of the
based on our recent articles1-3 updated with current monotarget therapies considering multitarget
data focusing on the concept of synergy, which has causes for these diseases is questioned.
c) Modern medical therapy presently uses more
_________________ and more combination therapies in the
*Correspondent author
treatment of several diseases like cancer,
Telephone +49 228 2872 2674
Fax +49 228 2872 2674 cardiovascular or rheumatic diseases often
E-mail: gudrun.ulrich-merzenich@ukb.uni-bonn.de without observing the originally expected
ULRICH-MERZENICH et al.: NATURAL PRODUCTS AND SYNERGISTIC EFFECTS 209

cummulation of adverse events (ADS) of both • Correspondingly, the overall effect with
single treatments. This has lead to an antagonistic interaction is less than expected from
increasing interest in the synergy concept3. the summation of the separate effects
d) Ancient systems of medicine like the (equation 2). A convex curve will be obtained.
Ayurvedic and Chinese system provide an • With the existence of a real synergism with
increasing amount of data demonstrating potentiated or over-additive effects, the overall
clinical efficacy in some of the known health effect of two drugs a and b that are applied
challenges. The term “reverse pharmacology” together as a mixture must be larger than it would
has been coined by Patwardhan et al.5 and is be expected by the summation of the separate
proposed to be a new path for drug discovery. effects. The result is then a concave curve
e) The pre-clinical development of synthetic (equation 3).
drugs has become extremely costly with costs
calculated around $ 400 million6. Equation 1: E (da, db) = E (da) + E (db)
f) The world wide demand for phyto- Equation 2: E (da, db) < E (da) + E (db)
pharmaceuticals grows steadily. In 2005 the Equation 3: E (da, db) > E (da) + E (db)
herbal industry had a turnover of about US$ 62
billion7. The world bank reports state that trade E stands for observed effect; da and db are the doses of
in medicinal plants, botanical drug products agents a and b.
and raw materials is growing at an annual In case of a synergism, lower amounts of agents a
growth rate between 5-15 % (Ref. 7). and b (doses=d) are necessary to achieve the effect.
g) It is rather agreed upon that nature’s The achieved synergy effect can amount to doubling
biodiversity has so far remained largely or even greater multiplication of the expected effect.
unexplored8 and thus still offers a tremendous Connected with the option of a dose reduction, it
potential. could be expected that at correctly chosen
h) The introduction of the “omic” – technologies combination of a natural product (a) with a strongly
as high-throughput methods opens the effective synthetic product (b), the potential of side
methodological possibility to investigate effects of agent (b) can be reduced simultaneously2.
complex mixtures like whole plant extracts. Mechanisms of synergy effects based on classical
They provide a new tool to investigate pharmacological, molecular biological and clinical work
polyvalent pharmacological activity, synergy can be divided into at least following four mechanisms2:
effects and thus multitarget treatments on a a) synergistic multitarget effects: natural products
rational basis. affect not only one target, but several targets
Definition of synergy from a pharmacological and can cooperate in an agonistic and
perspective synergistic way;
It is rather difficult to give an unequivocal
universal definition for the term synergy effect2. But
the “isobole method” of Berenbaum9-11 seems to be
one of the experimentally most convenient and also
the most demonstrative method among all those so far
proposed for the proof of synergy effects (Fig. 1); the
x and y axes reflect the dose rates of the single
individual components. Different dose combinations
are investigated for the same effect. An isobole is
understood to be a line or curve between points of the
same effect.

• According to Berenbaum9-11, the zero (0) – or


Fig. 1Isoboles for zero-interaction (effect-addition of individual
additive interaction means that the effect of two
components), synergism (positive interaction or potentiation) and
substances a and b is a pure summation effect antagonism (negative interaction) (reprint with permission,
(equation 1). Wagner et al.3)
210 INDIAN J EXP BIOL, MARCH 2010

b) pharmacokinetic and physiochemical effects: demonstrated multitarget effects. The term “omic”-
natural products improve solubility and/or the technologies relates to high-throughput technologies.
resorption rate and thereby the bioavailability; They are capable to determine a multitude of molecules
c) interference with the resistence mechanisms of on the gene (genomics, transcriptomics) or the protein
bacteria; natural product antagonize the level (proteomics, metabolomics) at the same time. They
resistence to antibiotics; and include different platforms of gene and protein
d) elimination and neutralization effects; natural microarrays. The commonly used gene micorarray
products by itself or after treatment (e.g. heating) platforms over the past 10 years like Affimetrix, Agilent
eliminate or neutralize within a drug preparation or PIQORTM allow to study the gene expression levels of
or in combination with synthetic drugs adverse the complete genome. The ability of these arrays to
events, so that altogether efficacy improves. simultaneously interrogate thousands of transcripts has
Examples for each mechanism are given in Table 1. led to important advances in a widerange of biological
A list of herbal drugs with evidences for synergistic problems including pharmacogenomic responses13.
effects or polyvalent activities has been given earlier1. Nevertheless, these techniques have certain limitiations
(e.g. background levels of hybridisation13). Thus,
Synergism in the context of the multitargeting sequencing based approaches measuring gene
approach expression have evolved. Ultra-high-throughput
Investigations of synergism as described by sequencing is emerging as an attractive alternative to
Beerenbaum9-11 (E(a,b)>Ea+Eb) have so far examined microarrays for genotyping, analysis of methylation
the combinatorial effects on one target at a single time patterns, the identification of transcription factors, but
point in one experimental setup. But the phenomenon also as platform to study mRNA-expression levels13.
of synergism of drug action in a human body or any First studies comparing the sequencing data from an
living organism is likely to be a dynamic process and Illumina sequencing platform with data obtained from a
a multitarget phenomenon. Imming et al.12 already “conventional” Affymetrix microarray revealed
described that it ultimately would be desirable to comparable results. But the ultra-high throughput
move away from a static to a dynamic target sequencing allowed for additional analysis like a
definition of drug action. detection of low-expressed genes, alternative splice
The developing “omic”-technologies provide us now variants, and novel transcripts13,14. It may be predicted
with the possibility to detect the interaction of a drug from this that the sequencing approach will be over time
with several targets and have indeed already widely adopted for measuring mRNA expression levels.
Table 1 Mechanisms of synergism (according to Wagner and Ulrich-Merzenich2)

Mechanism Example
Multitarget effects Cannabis46-48
Antispastic effects of Tetrahydrocannabinol + Cannabis extract (Cannabidiol increases the transport of
anandamide through the membrane)
Hypericum perforatum49
Hypericum, Hyperforin, Hypericin, Amentoflavon target different receptors in the pre- and
postsynaptic neuron
Pharmacokinetic effects Atropa belladonna50
Resorption of 1-Hyoscyamin is increased in the presence of flavonol-triglycosides
Ammi visnaga51
Khellin as part of the extract is better absorbed than Khellin alone.
Hypericum perforatum52
Hypericin combined with Polyphenols (Epicatechin, Procyanidin, Hyperosid, Rutin) enhances
bioavailability of Hypericin
Elimination of resistence Penicillin + clavulinic acid : antagonization of penicillinase54
mechanisms of bacteria53 ß-lactam antibiotics +Epigallocatechin gallate (EGCg) combined attack on the peptidoglycan part of
the bacterial wall55
Inhibition of lactam- or ester-cleaving enzymes by EGCg to maintain penicillin activity56
Norfloxacin, Amphotericin + Origanum vulgare, Pelargonium graveolens or Melaleuca alternifolia
⇒ reduction of the antibiotics through synergy57,58
Elimination, neutralisation Radix Aconiti
of toxically acting constituents Four methods to reduce toxic Aconitin to ~0.2%2
ULRICH-MERZENICH et al.: NATURAL PRODUCTS AND SYNERGISTIC EFFECTS 211

Nevertheless the “conventional” gene microarrays (ASA) and by the multiextract mixture Phytodolor
provided already new evidences in the context of (PD) as well as its single extract components using
synergy research and multitargeting3. E.g. the gene- the topic defined PIQUORTM Skinpatho Microarray
expression profile of α-tocopherol showed the which measures the modulation of inflammatory
targeting of genes related to the immune system, as genes. The multiextract mixture PD is composed of
well as the activation of genes related to the lipid three alcoholic extracts: Populus tremula (S1),
metabolism and to inflammation interestingly with no Solidago virgaurea (S2) and Fraxinus excelsior (S3).
significant change in the expression of classicial S1 possess a high content of salicin, salicylates* and
antioxidant genes15. Similarly methotrexate (MTX) or salicyl alcohols, but also phenolic components like
mecaptopurin targeted a multitude of genes involved flavonoids and catechins17. S2 has a high content of
in apoptosis, mismatch repair, cell cycle control and rutin and S3 a high content of fraxin. Each of the
the stress response16. This demonstrates convincingly single extracts modulated a different number of genes
that “single synthetic drugs or natural single based on the microarray assessment: S1 modulated 51
component/substances” have a multitude of targets on genes, S2 24 genes and S3 31 genes. The extract
the genetic level (Table 2). Simultaneous proteomic combination PD did not reveal an additive modulation
analysis demonstrates that at least part of the genomic of genes, it modulated “only” 40 genes and ASA 44
regulation is translated into proteins (Table 2). Thus, different genes. Thus, the number of active
the “omic” technologies lead us away from the components in an extract does not necessarily
paradigm of one drug, one target and one disease. determine the number of targets18.
If monosubstances have not only one, but multiple It was further observed that the gene expression
targets, a group of drugs or multicomponent mixtures profiles of the single extracts S1, S2 and S3 did not
are expected to have multiples of these targets. This allow a prediction of the gene expression profiles of
multiplicity of targets and the expected subsequent their combination (S1+S2+S3)18. A comparable
unpredictable mode of action has over many years been observation was made by Cheok et al.16 earlier. Gene
the main argument against the use of microarrays (HG-U95A oligonucleotide microarray)
phytopreparations. It was also expected that the adverse were applied to investigate the mode of action of
events (ADS) of single drugs would equally multiply in methotrexate (MTX) and mercaptopurin alone and in
a combination therapy. However, the potential of combination on human leukemia cells. It was
synergistic positive effects was not really considered. demonstrated that the combination of both in human
With respect to the multiplicity of targets, it has leukemia cells causes a different gene expression
been observed that the number of active components
________________
of a plant extract does not necessarily influence the *
salicylates: esters of salicin like salicortin, tremulacin or
number of targets. We compared in a fibroblast model 2’-O-acetylsalicortin. These prodrugs are metabolised like salicin
the modulation of genes by acetylsalicyclic acid to the active drug salicyclic acid.
Table 2  Monosubstance “Drugs” with multiple targets (modified according to Ulrich-Merzenich et al.3)

Drug/monosubstance Gene targets (examples) Protein-targets (examples)


α-Tocopherol Immune response, phosphate metabolism, protein Proteinkinase C, Phospholipase A2 5-Lipoxy genase 3
modification, lipid metabolism, inflammatory
responses15
Daidzein Androgenresponsive genes, IGF-1 Pathway, Oestrogen-receptors, Peroxisome-Proliferator-Activated
MAP kinase pathway genes63 Receptors59
Quercetin Immune Response (Interleukines like IL-1, IL-6 20 protein (e.g. heat shock proteins, annexins involved in
or cytokines like TNF-α) 3 apoptosis, metabolism, detoxification and gene regulation60
Epigallocatechin-3- Zinc finger protein 33A, Ubiquitin-activating Mitogen-activated protein kinases, nuclear factor-κB and
gallate enzyme E1C, basic fibroblast growth factor in others61,62
human bronchial epithelial 21BES cells62
Methotrexate S 100 calcium binding protein A8 and A9; Dihydrofolic acid, T-Cells, Livertransaminases (Folic acid
Leucocyte immunoglobulin receptor; densin, synthesis inhibition)
alpha 3, neutrophil specific16
Mercaptopurin Splicing factor proline/glutamine rich; nucleoside Adenylosuccinate-synthase, Phosphoribosylpyrophosphat-
diphosphatase kinase type 6, CDC28 protein amidotransferase, (DNA- and RNA-synthesisinhibition)
kinase 2, polymerase (DNA directed),gamma 216
212 INDIAN J EXP BIOL, MARCH 2010

profile than each single applied drug. The expression be essential to demonstrate that single plant
profile of the combination shared only 14% of genes ingredients or plant extracts yield reproducible and
with the ones of the single application. Thus, 86% of specific expression profiles (signatures) byusing the
the previous modulated genes did not respond16. This array technology. Figure 2 shows the gene expression
demonstrates that drug combinations can lead to the profile of the willow bark extract STW-33-1 in human
activation of more or less entirely different sets of chondrocyte cultures. In four different microarrays
genes compared to those activated by the single (Agilent human whole genome microarray), a specific
agents1,3. Thus, the application of drug combinations profile of the willow bark extract in comparison to
including phytopharmaceuticals does not necessarily other tested substances could be recognized. In the
mean the addition or multiplication of targets. It can given example four arrays with two different
lead to new modes of actions. The consequences of concentrations of the willow bark extract were
these findings specifically for phytopharmaceuticals investigated. For standardisation, however, the use of
are: the mode of action of a medication is most
reliably assessed if all applied single drugs are taken
into consideration since a new mode of action may
arise through the combination of drugs. Therefore, the
demand to demonstrate the mode of action of each
single component in a phytopharmaceutical may not
be obligatory any more.
The “omic”- approach as methodology for the
development of plant drugs
Standardization of single and multicomponent
plant extracts—Prerequisite of the use of
phyotpharmaceuticals is their standardisation. This
starts with the proof of the authenticity and quality of
the plant material, which has for long relied on
morphological and chemical methods. Metabolomic
fingerprinting by GC-MS, HPLC-MS or NMR-
spectroscopy are increasingly used with the aim to
develop metabolomic profiles. The principle
component analysis (PCA), which is used to reduce
the complexity of spectral data has simplified the
visualisation (and recognition) of profiles from the
multiple data sets. Daniel et al.19 proposed to use PCA
besides standard analytic methods for the
characterisation and identification of plants and
multicomponent plant extracts. DNA-based assays
have also been introduced for complementing the
morphological and chemical methods with the aim to
generate molecular “bar codes” for the correct
identification of medicinal plants20. Demands are
increasing to establish open access data banks in an
multidisciplinary effort for medicinal plant research
Fig. 2The hierarchically clustered map of genes expressed in
and to include e. g. the obtained data (profiles) into human chondrocytes stimulated with willow bark extract (STW-
the European Pharmacoepia in order to develop a 33-1, lines 1,2: 50 µg/ml, lines 3,4: 30 µg/ml), quercetin (lines
common standard for the characterisation of plants. 4,6: 5, 10 µM), diclofenac (line 7: 30 µg/ml) and (acetylsalicyclic
Another level of standardisation lays on the acid, line 8: 50 µg/ml) for 6 h. The figure shows only a small part
of the clustered gene map exemplatory. A red colour indicates the
assessment of a reproducable mode of action of plant up-regulation of a gene (names of the genes are indicated on the
ingredients or plant extract irrespective of a certain right side of the map), the green colour indicates the down-
degree of variation in their composition. Here it will regulation of a gene. (Ulrich-Merzenich et al.3)
ULRICH-MERZENICH et al.: NATURAL PRODUCTS AND SYNERGISTIC EFFECTS 213

more than four microarrays in different concentration The finding that reproducible gene expression
would be more reliable to obtain the specific and profiles can be obtained in cell culture models for
reproducable core set of genes, which is activated by a plant extracts is supported by recent findings of
plant extract. Lines 7 and 8 show the gene expression Pakalapati et al.21. Rats were treated with a Trifolium
profile of the reference substances diclofenac and pratense extract rich in isoflavones. Whole genome
acetylsalicyclic acid (ASA). The comparison of microarray (Affymetrix Rae 230_2) measurements
expression profiles of complex mixtures with known showed a reproducibly modulated core set of genes.
reference substances (monosubstances) is useful, since The microarray data were combined with proteom
this will simplify the identification of the essential studies. Those revealed that the modulation of several
pathways which finally lead to the observed clinical or genes of the lipid metabolism was converted onto the
pharmacological effects. Figure 3 shows the hierachical protein level.
clustering of genes which are simultaneously up- or The reproducibility of metabolomic expression
down regulated. Thereby common pathways are profiles induced by multicomponent mixtures is
identified. In order to cope with the flood of already supported by several studies in the field of
bioinformatic data, it will be advisable not to start nutrigenomics, e. g. Wang et al.22 have demonstrated
measurements with too complex combinations of with an 1H NMR spectroscopy-based study in
extract mixtures. conjunction with chemometric methods differences

Fig. 3The hierarchically clustered map of genes expressed in blood cells of rats after treating the animals with different fractions of the
aequeous willow bark extract (STW-33-1). Each line represents the blood cell expression profile of one animal. (Line 1-3: controls; line
4-6: ethanol fraction; line 7-9: ethylacetat fraction; line 10, 11: watery fraction; line 12: Total extract (TS)). The figure shows only a small
part of the clustered gene map as illustration. A red colour indicates the up-regulation of a gene (names of the genes are indicated on the
right side of the map), the green colour indicates the down-regulation of a gene. (Ulrich-Merzenich et al.3)
214 INDIAN J EXP BIOL, MARCH 2010

between the metabolome pattern induced by UCSC Genome Browser, Esembl. and non-sequence
chamomile tea in urine before and after tea ingestions. centric databases like the Protein Data Bank (PDB)
The effect of a dietary intervention of soy isoflavones has been given by Baxevanis28.
in humans was equally detectable by NMR23, 24. Pharmacokinetics and bioavailability of plant
For a complete understanding of the mode of extracts and their combinations–Pharmacokinetic and
action, the integration and correlation of data sets bioavailability studies are an essential need to
from the different “omic” areas and from functional determine the exact pharmacological action of
tests is essential, but on the other hand tremendously phytopharmaceuticals, but still unsufficient data do
challenging. For example, not all expressed mRNAs exist. Due to the high number of components in
are converted into proteins, the mRNA-splicing herbal drugs, their variable absorption and their
process needs to be considered. Also the differential complex biotransformation, assesments with complete
timings of these events is a challenging factor. coverage have been practically impossible by
Bilello25 proposed for the understanding of conventional methods. The new high-throughput
pathophysiological processes an integrated system technologies make these kinds of assessments
based approach involving modelling and simulating possible and will improve the speed and yield. But
the complex dynamic interactions between genes, again, even after the identification of the available
transcripts, proteins, metabolites and cells, plant components and their metabolites in plasma,
encompassing many of the “omic” technologies and functional studies are essential for determining the
using computational and mathematical models to mode of action. This includes toxicology testing – the
analyse and simulate networks and pathways. This latter being the best developed field in the context of
could also take into account spatial and temporal “omic”-technologies so far.
relationships that give rise to cause and effect in Assesment of the toxicity of plant extracts—It is
biological systems25. Data of such integrated generally expected that the use of gene expression
approaches have so far not been published in the field data is more sensitive than traditional toxicological
of phytomedicine. However, an integrated approach endpoints29. Searfoss et al.29 identified the following
based on “omic” technologies was already used e.g. general goals for the new field of Toxicogenomics
by Kleno et al.26, who identified potential biomarkers equally applicable to the development of synthetic
(related to the glucose and lipid metabolism and to drugs or phytopharmaceuticals: a) understand
oxidative stress) for hepatotoxicity in rats. Rochfort24 mechanisms of toxicity; b) predict toxicity; c) develop
showed that correlations from liver mRNA, liver in vivo and in vitro surrogate models and screens; and
proteome, and metabolome analysis of serum d) develop toxicity biomarkers. These should lead to
corresponded to changes in glucose, lipid metabolism an improvement of safety, to the shortening of the
and oxidative stress responses. Mayr et al.27 combined drug development and a cost reduction. Recent
proteomic and metabolomic studies in the reviews on the issue have been published by Searfoss
cardiovascular system and described the results that et al.29, Suter et al.30 and Stork et al.31.
“the simultaneous assessment of protein and There is already a collaborative effort between the
metabolite changes translated purely descriptive European Molecular Biology Laboratory-European
proteomic and metabolomic profiles into a functional Bioinformatics Institute ArrayExpress, the
context and provided important insights into International Life Sciences Institute Health and
pathophysiological mechanisms that would not have Environmental Science Institute, and the National
been obtained by other techniques.” Institute of Environments Health Sciences National
For standardisation of plant extracts, the primary Centre for Toxigenomics Chemical Effects in
focus will be the reproducibility of profiles with the Biological Systems knowledge base to establish a
different “omic”-techniques resulting most likely in a public infrastructure on an international scale and
tremendously improved insight into the examine other developments aimed at establishing
pharmacological mode of action. Issues like the toxicogenomics data-bases. An overview of the major
permitted magnitude of variation for each technology, toxicogenomics database efforts in the public sector
however, still need to be defined. An overview about has been given by Mattes et al. 32. These data bases
the currently available biological databases for are under development and offer on one hand
annotations of genes and proteins such as GenBank, information (gene expression profiles) on the toxicity
ULRICH-MERZENICH et al.: NATURAL PRODUCTS AND SYNERGISTIC EFFECTS 215

endpoints dependant on organs as well as on different so called off- and on-target toxicities early in the drug
diseases. For example, it has been estimated that the development. Different tissues or cells can be exposed
liver has 10-20 distinct toxic phenotypes, including to the selected plant extracts and then screened for
steatosis, multifocal necrosis and hypertrophy29,33. On mRNA or metabolite modulation relevant for toxicity.
the other hand they accept submissions from Off-target toxicities are those caused by actions un-
investigators to create and increase these data bases. related to the targets such as non-specific immune or
The Chemical Effects in Biological Systems inflammatory reactions or hepatotoxicities due to drug
Knowldege Base (CEBS) data base metabolism effects29. On-target toxicities are
(http://www.niehs.nih.gov/nct/cebs) is under additional unintendent effects due to the interaction
development. Specific objectives have been given by of target and drug. Besides these already existing
Mattes et al.32 as follows: a) to compare products (Table 3), it will be necessary to develop
toxicogenomic effects of chemicals/stressors across additional gene- or proteinchips specifically suited to
species – yielding signatures of altered molecular the toxicology demands of phytopharmaceuticals.
expression; b) to phenotypically anchor these changes Synergism, signal cascades and trigger points
with conventional toxicology data – classifying
of the metabolism—Synergism can occur through
biological effects as well as disease phenotypes; and
multitargeting. But the identification of multiple
c) to delineate global changes as adaptive,
targets, which will be simplified through the use of
pharmacologic or toxic outcomes – defining early
“omic” techniques, will be only one part of the puzzle
biomarkers, the sequence of key events and
to understand synergistic actions. The type of target,
mechanism of toxicant actions. CEBS is a dynamic
which can be as diverse as enzymes, receptors,
concept for integrating large volumes of
antibodies or signalcascades is equally important.
transcriptomic, proteomic, metabonomic, and
Signalcascades are highly interesting targets for drug
toxicological knowledge in a framework that serves as
development since their activation can lead to an
a continually changing «heuristic» engine32. For
amplification of the original signal by a million times.
further information and internet adress lists see Mattes
The investigation of such signal cascades may
et al.32.
elucidate our understanding of synergistic, but also
Some predictive toxicology products have already
antagonistic effects.
emerged by different biotechnology firms (Table 3).
The development of these microarrays is based on the Recently G-protein-coupled receptors (GPCRs)
above mentioned type of investigations of gene have gained increasing importance as targets for drug
expression signatures or “fingerprints” which are development34. GPCRs constitute a large family of
highly predictive for toxicity endpoints (phenotypes)29 receptors which initiate various signal cascades. Such
by testing toxic (test) compounds. These products can cascades may be called trigger points of the
be readily used for phytopharmaceuticals to screen for metabolism, not only because they amplify a signal,
Table 3Predictive toxicology products (according to Searfoss et al.29 and data of Ulrich-Merzenich et al.3)

Provider Product Description of Product/approach


Iconix.Pharmaceuticals DrugMatrix™ Chemogenomics Global expression profiling in vivo+in vitro
www.iconixpharm.com >600 compounds chemogenomic database,
+200 validated gene signature sets
Gene Logic www.genelogic.com ToxExpress™ ToxExpress™ reference data base predictive
system with in in vivo+in vitro models.
Extensive liver toxicity database.
Exonhit www.exonhit.com Sale-Hit™ DATAS™ screening technology reveals the
presence of alternatively spliced mRNAs, and
thus protein isoforms linked with toxicity
Miltenyi Biotec Piquor Tox Microarray Determination of classical toxic endpoint genes
www.miltenyibiotec.com relating to apoptosis, DNA-damage + repair,
inflammation, ox-stress
Affymetrix www.affymetrix.com GeneChip® Rat Tox U34 850 mRNA transcripts and EST clusters
selected from UNIGene database (Build 34)
Agilent www.agilent.com/chem Agilent 7890A/NPD/5975C/DRS/GC/MSD Forensic Toxicology Data Base of 277
compounds
216 INDIAN J EXP BIOL, MARCH 2010

but they often can divert the signal into opposing them. The potential of curcumin to modulate
directions. One example of such a trigger point is the signalling pathways leading to cell cycle regulation or
mitogen activated protein kinase (MAPkinase) directly altering cell cycle regulatory molecules in
cascade. Several molecules like H2O2, vitamin C or cancer therapy is presently under intensive
the cytokine TNF-α35,36 can activate this cascade. The investigation44. Synergism as the result of the
result of activation can be twofold-apoptosis or coordinated activation of signal cascades may be a
proliferation. E.g. the stimulation with a low dose of highly rewarding area for drug development.
H2O2 over a short time leads to the activation of the Synergism as result of an activation of transport
MAPkinases ERK-1 and 2 and mounts into proteins to increase the bioavailability of certain drugs
proliferation. Contrary, a long time exposure of H2O2 will be described in a future review.
in low dosage leads to apoptosis. A high dosage
shortly given or applied for a long time leads also to New combinations of plant components based on
apoptosis3,36,37. Thus, the activation of the same signal traditional systems of medicine
cascade can lead to opposite effects depending on Traditional systems of medicine like the Chinese or
time and dosage of the trigger. Higher dosages of Ayurvedic systems have detailed description of
H2O2 activate besides ERK-1 and 2 also the so called methods to prepare their medicines. Even though
“stress induced” MAPkinases p38 and c-jun. Their tremendous efforts have been made e.g. in India over
additional activation leads to apoptosis instead of the past 40 years supported by the Council of
proliferation36,37. Research in Ayurveda, Siddha and Unani to
standardize these methods and at the same time to
Such dosage dependent reversal effects have been justify the different procedures, both goals are
observed for several herbal drugs like Ginkgo difficult to achieve due to the often high number of
biloba38, Hypericum perforatum39 or Curcuma plants in a single medication and due to the high
longa40. One possible explanation for reversal, but number of steps involved in the preparation. As
also synergistic effects could be such an activation of already mentioned the “omic”-technologies allow the
signal cascades. In case of reversal effects, the assessment of complex mixtures and could be used to
signalcascades antogonize each other in certain monitor the different steps of the preparation in order
dosage ranges; in case of synergism signalcascades to validate or dismiss the original recipes and to look
interact and agonize each other. into the mode of action of Ayurvedic preparations
Indeed, the ultimate result of the activation of a which contain sometimes up to 70 different plants45.
signal cascade can be due to signal amplification Chinese medicines are primarily prepared as
much greater than originally calculated by the watery decoctions. It is recognized since long that this
summation of the single effects. Imming et al.12 called type of preparation is difficult to standardize. The
these effects vaguely “rippels” in drug action. The Chinese Pharmacopeia is based on the analyses of
“omic”-technologies can support the analysis of the alcohol extracts. The results, however, may not mirror
simultaneous up- or downregulation of signal cascade the contents of the decotions. Figure 2 shows that the
molecules and thereby may factually describe these application of different preparations of plant extracts
“rippels”. Gene microarray-analysis identify via in rats lead to clearly distinguishable patterns of gene
hierarchical clustering genes which are commonly up- expressions in blood cells. These differences are
or down regulated (Fig. 2). Thereby pathways are likely to be translated onto the protein and functional
identified. Synergism may be the result of the level, too. Thus, “omic” technologies can be used to
activation of central trigger points of such pathways support the selection and development of the optimal
and may be an expression of a nonlinear correlation of extract form or preparation for Chinese or Ayurvedic
molecular interactions (of such pathways) and clinical medications in an acceptable time frame. However,
effects. Such trigger points could be taken as new the following premises and recommendations may be
drug target or as targets to be protected to safeguard a considered to obtain reliable evidence-based “new”
normal metabolism. As already mentioned the GPCRs phytopharmaceuticals:
have gained recently an extremely high interest as 1. The basic material should be quality assured in
drug targets34. Several plant constituents like form of a standardization according to conventional
genistein41 or flavonoids42 like naringin and cyanidin- methods (fingerprint and if available fixed content of
3-glucosides43 were already shown to interact with biological active substances).
ULRICH-MERZENICH et al.: NATURAL PRODUCTS AND SYNERGISTIC EFFECTS 217

2. A toxicology screening of the “phyto- and (d) in the analysis of the mode of action of plants
pharmaceutical” should be performed with one of the and multiextract mixtures; should create in a long
presently already available screening tools (Table 3). term perspective evidence-based plant medications
In this context specific toxicology assays and data and should lead to the discovery of effective
banks for critical plant components need to be phytomedical interventions in the prevention and
developed. treatment of different diseases in form of new single
3. An evaluation of the possibility to identify treatments or new combinatory drug regimes while
“surrogate” plant components which represent the exploiting synergy-effects. However, presently we are
activity of the plant extract should follow and could far ahead with data collection, but still without a
be used for standardization. thorough conceptual framework for the integration of
4. Standardisation of the plant extracts according to the tremendous amount of data. System biology is
an expression profile taking into consideration evolving as a key discipline to address this task,
relevant concentrations and the bioavailability. which is a multidisciplinary challenge still based on
Matching of plant signatures with signatures of the experimental biology.
mode of action in different surrogate models should
be attempted. References
5. Determination of the down-stream effects 1 Ulrich-Merzenich G, Zeitler H, Panek D, Vetter H & Wagner
influenced by phytopharmaceuticals and bioactive H, Application of the„Omic“-technologies in phytomedicine,
Phytomedicine, 14/1 (2007) 70.
plant components via identification of molecular 2 Wagner H & Ulrich-Merzenich G, Synergy research:
pathways including the identification of relevant Approaching a new generation of phytopharmaceuticals Part
mechanisms of action for the different plant I, Phytomedicine, 16/2-3 (2009) 97.
components through comparison with reference-listed 3 Ulrich-Merzenich, Panek D, Zeitler H, Wagner H & Vetter
H, New perspectives for synergy research with the „omic“-
drugs. technologies, Phytomedicine, 16/6 (2009) 495.
6. Integration of the profiles of single plant 4 Cragg G M & Boyd M, Drug discovery and development at
components and complex extract mixtures into the the National Cancer Institute, in Medicinal Plant Resources
international freely accessible data banks of genetic, of the Tropical Forest edited by M J Balick, E Elisabetsky, S
proteomic and metabolomic screens. A Laird (Columbia University Press, New York) 1996, 101.
5 Patwardhan B & Mashelkar R A, Traditional medicine-
7. Exploring the connectivity of biological insprired approaches to drug discovery: can Ayuveda show
pathways by a systematic combination screening and the way forward? Drug Discov Today, 14 (2009) 804.
thereby determining the mode of action (antagonistic, 6 Vernon J A, Golec J H & Dimasi J A, Drug development
additive, agonistic=synergistic) of single components costs when financial risk is measured using the Fama-French
three-factor model, Health Econ, Aug 4 (2009) Epub ahead
and extract mixtures in comparison to reference-listed of print.
drugs. Complex mode of actions of combination 7 Patwardhan B, Warude D, Pushpnagadan P & Bhatt N,
therapies (Table. 2) should be investigated. Ayurveda and traditional Chinese medicine: a comparative
8. Based on plant and disease profiles relevant overview, Evid Based Complement Alternat Med, 2 (4)
clinical applications for phytopharmaceuticals should (2005) 465.
8 Pieters L & Vlietinck A J. Bioguided isolation of
be identified. pharmacologically active plant components, still a valuable
strategy for the finding of new lead components? J
Outlook and conclusions Ethnopharmacol, 100 (2005) 57.
In future drug development from natural products 9 Berenbaum M C, Synergy, additivism and antagonism in
will not necessarily rely only on the discovery and immunosuppression, a critical review, Clin Exp Immunol, 28
analysis of new structures from nature’s extremely (1977) 1.
10 Berenbaum M C, Criteria for analysing interactions between
rich biodiversity, but can systematically explore biologically active agents, Adv Cancer Res, 35 (1981) 265.
combinatory drug regimes. The introduction of the 11 Berenbaum M C, What is synergy? Pharmacol Rev, 41
high throughput-technologies makes this possible. (1989) 93.
Their use has already changed our conceptual frame 12 Imming P, Sinning C H & Meyer A, Drugs, their targets and
work of drug targets, drug action and our the nature and number of drug targets, Drug Discov, 5 (2006)
821.
understanding of the pathology of diseases 13 Marioni J C, Mason C E, Mane S M, Stephens M & Gilad Y,
(in favour of using phytomedicine). The systematic RNA-seq: An assessment of technical reproducibility and
application of these technologies in the assessment of comparison with gene expression arrays, Genome Res, 18
(a) the authenticity; (b) the quality; (c) the toxicology; (2008) 1509.
218 INDIAN J EXP BIOL, MARCH 2010

14 Van Iterson M, t’Hoen P A C, Pedotti P, Hooiveld G J E J, den 31 Storck T, Von Brevern M C, Behrens C K, Scheel J & Bach
Dunnen J T, van Ommen G J B, Boer J M & Menezes R X, A, Transcriptomics in predictive toxicology, Curr Opin Drug
Relative Power and sample size analysis on gene expression Disc Devel, 5 (2002) 90.
profiling data, BMC Genomics, 10 (2009) 439 (e-pub). 32 Mattes W B, Petti S D, Sansone S A, Bushel P R & Waters
15 Vasu V T, Hobson B, Gohil K & Cross C E, Genome-wide M D, Database development in toxidogenomics: Issues and
screening of alpha-tocopherol sensitive genes in heart tissue Efforts, Environ Health Perspect, 112/4 (2004) 495.
from alpha-tocopherol transfer protein null mice (ATTP(-/-)), 33 Zimmerman H J, Drug-induced liver disease, Clin Liv Dis,
FEBS Lett, 581 (2007) 1572. 4/1 (2000) 73.
16 Cheok M H, Yang W, Pui C H, Downing J R, Cheng C, 34 Heilker R, Wolff M, Tautermann C S & Bieler M, G-protein
Naeve C W Relling M V & Evans W E, Treatment-specific coupled receptor-focused drug discovery using a target class
changes in gene expression discriminate in vivo drug platform approach, Drug Discov Today, 14/4-5 (2009) 231.
response in human leukemia cells, Nat Genet, 34 (2003) 85. 35 Karin M, Inflammation-activated Protein Kinases as targets
17 Nahrstedt A, Schmidt M, Jäggi R, Metz J & Khayyal T, for Drug Development, Proc Am Thorac Soc, 2 (2005) 386.
Willow bark extract: the contribution of polyphenols to the 36 Ulrich-Merzenich G, Zeitler H, Panek D, Bokemeyer D &
overall effects, Wien Med Wochenschr, 157/13-14 (2007) 348. Vetter H, Vitamin C promotes human endothelial cell growth
18 Ulrich-Merzenich G, Jobst D, Zeitler H, Müller J & Vetter H, via ERK-signaling pathway, Eur J Nutr, 46 (2007) 87.
Measurement of synergistic effects of a phytopharmaceutical 37 Ulrich-Merzenich G, Zeitler H, Vetter H & Kraft K, Synergy
by microarray-analysis, Planta medica, 73 (2007) SL026. research: vitamins and secondary plant components in the
19 Daniel C, Kersten T, Kehraus S, König GM & Knöß W. maintenance of the redox-homeostasis and in cell signaling,
Identifizierung und Charakterisierung von Arzneipflanzen Phytomedicine, 16/1 (2009) 2.
mit “Metabolic Fingerprinting”, Z Phytotherapie, 29 (2008) 38 Chan W H & Hsuuw Y D, Dosage effects of ginkgolide B on
270. ethanol-induced cell death in human hepatoma G2 cells, Ann
20 Sucher N J & Carles M C, Genome-based approaches to the N Y Acad Scie, 1095 (2007) 388.
authentication of medicinal plant, Planta medica, 74/6 39 Franklin M, Hafizi S, Reed A, Hockney R & Murck H, Effect
(2008) 603. of sub-chronic treatment with Jarsin (extract of St. John’s wort,
21 Pakalapati G, Li L, Gretz N, Koch E & Wink M, Influence of Hypericum perforatum) at two dose levels on evening salivary
red clover (Trifolium pratense) Isoflavones on gene and melatonin and cortisol concentrations in healthy male
protein expression profiles in liver of ovariectomized rats, volunteers, Pharmacopsychiatry, 39/1 (2006) 13.
Phytomedicine, 16/9 (2009) 845. 40 Ali R E & Rattan S I, Curcumin’s biphasic hormetic
22 Wang Y, Tang H, Nicholson J K, Hylands P J, Sampson J & response on proteasome activity and heat-shock protein
Holmes E, A metabonic strategy for the detection of the synthesis in human keratinocytes, Ann N Y Acad Sci, 1067
metabolic of chamomile (Matricaria recutita L) ingestion, J (2006) 394.
Agric Food Chem, 53/2 (2005) 191. 41 Si H & Liu D, Phytochemical genistein in the regulation of
23 Solanky K S, Bailey N J, Beckwith-Hall B M, Davis A, vascular function: new insights. Curr Med Chem, 14/24
Bingham S, Holmes E, Nicholson J K & Cassidy A, (2007) 2581.
Application of biofluid 1H nuclear magnetic resonance-based 42 Spencer J P, Food for thought: the role of dietary flavonoids
metabonic techniques for the analysis of the biochemical in enhancing memory, learing and neuro-cognitive
effects of dietary isoflavones on human plasma profile, Anal performance, Proc Nutr Soc, 67/2 (2008) 238.
Biochem, 323 (2003) 197. 43 Dallas C, Gerbi A, Guillaume T, Juchaux F & Bernard F X,
24 Rochfort S, Metabolomics reviewed: a new “omics” platform Lipolytic effect of a polyphenolic citrus dry extract of red
technology for system biology and implications for natural orange, grapefruit, orange (SINETROL) in human body fat
products research, J Nat Prod, 68/12 (2005) 1813. adipocytes. Mechanism of action by inhibition of camp-
25 Bilello J A, The agony and ecstasy of “OMIC” technologies phosphodiesterase (PDE), Phytomedicine, 15 (2008) 783.
in drug development, Curr Mol Med, 5/1 (2005) 39. 44 Sa G & Das T, Anti cancer effects of curcumin: cycle of life
26 Klenø T G, Kiehr B, Baunsgaard D & Sidelmann U G, and death, Cell Division, 3 (2008) 1.
Combination of “omics” data to investigate the 45 Government of India, Ministry of Health and Family Planning,
mechanism(s) of hydrazine-induced hepatotoxicity in rats Department of Health, The Ayurvedic Formulary of India, Part
and to identify potential biomarkers, Biomarkers, 9/2 (2004) I, (Government of India Press) Faridabad 1978, 72.
116. 46 Zuardi A W, Shirakawa I, Finkelfarb E & Karniol I G,
27 Mayr M, Mayr U, Chung Y L, Yin X, Griffiths J R & Yu Q, Action of cannabidiol on the anxiety and other effects
Vascular proteomics: linking proteomic and metabolomic produced by ∆9–THC in normal subjects, Psychopharmacol,
changes, Proteomics, 4 (2004) 3751. 76 (1982) 245.
28 Baxevanis A D, The importance of biological databases in 47 Williamon E M & Evans F J, Cannabinoids in clinical
biological discovery, Curr Protoc Bioinformatic 1 (2006) practice, Drugs, 60/6 (2000) 1305.
Unit 1.1. 48 Wilkinson J D, Whalley B J, Baker D, Pryce G, Gibbons G,
29 Searfoss G H, Timothy P, Jolly R & Jolly R A, The role of Constanti A & Williamson E M, Medicinal Cannabis: is ∆9 –
transcriptome analysis in pre-clinical toxicology, Curr Mol THC responsible for all its effects, J Pharm Pharmacol,
Med, 5 (2004) 53. 55/12 (2003) 1687.
30 Suter L, Babiss L E & Wheeldon E B, Toxicogenomics in 49 Simmen U, Higelin J, Berger-Buter K, Schaffner W &
predictive toxicology in drug development, Chem Biol, 11 Lundstrom K, Neurochemical Studies with St. John’s Wort
(2004) 161. in vitro, Pharmacopsychiatry, 34 Suppl 1 (2001) 137.
ULRICH-MERZENICH et al.: NATURAL PRODUCTS AND SYNERGISTIC EFFECTS 219

50 List P H, Schmid W & Weil E, Reinsubstanz oder galenische 58 Rosato A, Vitali C, Gallo D, Millillo M A & Mallamaci R,
Zubereitung, Arzneim Forsch, 19 (1969) 181. The inhibition of Candida species by selected oils and their
51 Eder M & Mehnert W, Pflanzliche Begleitstoffe – wertvolle synergism with Amphoterin B, Phytomedicine, 15 (2008)
Hilfsstoffe oder überflüssiger Ballast? Pharm Unserer Zeit, 635.
29 (2000) 377. 59 Dang Z & Löwik C W, The balance between concurrent
52 Butterweck V, Jürgenliek-Wolff K G, Nahrstedt A & activation of ERs and PPARs determines daidzein-induced
Winterhoff H, Flavonoids from Hypericum perforatum show osteogenesis and adipogenesis, J Bone Miner Res, 19/5
antidepressant activity in the forced swimming test, Planta (2004) 853.
Med, 68 (2003) 577. 60 Wenzel U, Herzog A, Kuntz S & Daniel H, Protein
53 Hemaiswarya S, Kruthiventi A K & Doble M, Synergism expression profiling identifies molecular targets of quercetin
between natural products and antibiotics against infectious as a major dietary flavonoid in human colon cancer cells,
diseases, Phytomedicine, 15(8) (2008) 639. Proteomics, 4/7 (2004) 2160.
54 Lee N, Yuen K Y & Kumana C R, Clinical role of ß-Lactam/ß- 61 Chung J Y, Park J O, Phyu H, Dong Z & Yang C S,
lactamase inhibitor combinations, Drugs, 63 (2003) 15111. Mechanisms of inhibition of the RAS-MAP kinase signaling
55 Zhao W H, Hu Z Q, Okuba S, Hara Y & Shimamura T, pathway in 30.7b Ras 12 cells by tea polyphenols (-)-
Mechanism of synergy between epigallocatechin gallate and epigallocatechin-3-gallate and theaflavin-3,3’-digallates,
ß-lactams against methicillin-resistent Staphylococcus FASEB J, 15 (2001) 2022.
aureus, Antimicrob Agents Chemother, 45 (2001) 1737. 62 Vittal R, Selvanayagam Z E, Sun Y, Hong J, Liu F, Chin K
56 Zhao W H, Hu Z Q, Hara Y & Shimamura T, Inhibition of V & Yang C S, Gene expression changes induced by green
penicillinase by epigallocatechin-gallate resulting in tea polyphenol (-)-epigallocatechin-3-gallate in human
restoration of antibacterial activity of penicillin against bronchial epithelial 21BES cells analyzed by DNA
penicillinase producing Staphylococcus aureaus, Antimicrob microarray, Mol Cancer Ther, 3/9 (2004) 1091.
Agenst Chemother, 36 (2002) 2266. 63 Takahashi Y, Lagigne J A, Hursting S P, Chandramouli G V,
57 Rosato A, Vitali C, Laurentis N, Armanise D & Nulillo M A, Perkins S N, Kim Y S & Wang T, Molecular signature of
Antibactirial effect of some essential oils administered alone soy-derived phytochemicals in androgen-responsive prostate
or in combination with Norfloxacin, Phytomedicine, 14 cancer cells: a comparison study using DNA microarray, Mol
(2007) 727. Carcinog, 45/12 (2006) 943.

Você também pode gostar