Abbreviations
Abbreviations
CFU: colony forming units
FIC: fractional inhibitory concentration
GC-MS: gas chromatography coupled to mass spectrometry
MIC: minimum inhibitory concentration
Introduction
Introduction
The therapeutic value of synergistic interactions has been known since antiquity,
and many different cultural healing systems have relied on this principle in the belief
that combination therapy may enhance efficacy. The ancient texts pertaining to Ayurveda
and traditional Chinese herbal medicine describe formulas consisting of complex herbal
mixtures which may contain several plant-based ingredients [1]. African traditional healers rarely rely on a single plant for therapeutic regimens
but often combine various plant parts and different species in order to achieve optimal
results. The fundamental principle of aromatherapy is the combination of highly complex
different essential oils to achieve a therapeutic effect. The historical use and application
of polyherbals has been carried down through the centuries, and today allopathic medicine
commonly uses the very same principles to combine various molecules in single or separate
dosage forms which are administered concomitantly. Recently, the application of combination
therapy has gained a wider acceptance, especially in the treatment of infectious diseases.
The World Health Organization, for example, has urged pharmaceutical companies to
stop promoting the use of artemisinin derivatives in monotherapy. Instead, artemisinin
combination therapy should be encouraged not only because it has a cure rate of 95 %
against the malaria parasite (Plasmodium falciparum) but may also contribute to curb resistance. Multidrug therapy has become of paramount
importance in the fight against multidrug resistant microbial strains. Without the
current multidrug approach used to treat tuberculosis (isoniazid, rifampicin, pyrazinamide,
and ethambutol), the mortality of infected patients could reach global epidemic proportions.
Another renowned antimicrobial agent having a significant synergistic effect in combination
is amoxicillin (a β-lactam antibiotic) and clavulanic acid. Clavulanic acid binds to β-lactamase producing microorganisms, which protects amoxicillin from β-lactamase attack, which in turn results in an extended spectrum of activity for amoxicillin.
The concept of antimicrobial synergy is based on the principle that, in combination,
the formulation may enhance efficacy, reduce toxicity, decrease adverse side effects,
increase bioavailability, lower the dose and reduce the advance of antimicrobial resistance
[2], [3], [4]. New antimicrobial combination drugs which include natural product combinations
have recently become a research priority. This approach has financial implications
as reformulation of existing drugs or combinations may prove to be a more viable option,
rather than developing a new drug which will require extensive clinical trials for
verification. Furthermore, the ban imposed by the United States of America and European
Union on the use of allopathic antimicrobials in livestock farming has led to the
search for natural antimicrobial combinations that may impact positively on agricultural
and livestock farming [4].
Even though the concept of interactive antimicrobial therapy is well practised in
western medicine, and many anecdotal accounts of plants used in combination for the
treatment of microbe-related infections are evident, the validation of this phenomenon
in the field of pharmacognosy has been neglected. Medicinal plants offer a vast resource
of natural compounds, and the exploration of the various levels of interaction that
may exist include: constituents within a plant, interactions between different parts
of a plant, between different plant species, or the interaction with non-plant-based
antimicrobials. It is encouraging to observe that there has been a recent increase
in the number of publications reporting on plant-based pharmacological interactions.
The concept of synergistic principles from a pharmacological and/or phytotherapeutic
perspective has been addressed in various reviews [5], [6], [7], [8], but these have not specifically focused on the antimicrobial interactions. The
literature on the proposed methods appears fragmented and often confusing, and several
experimental designs have been proposed, which leaves the results inconclusive. This
review intends to succinctly collate the available literature which has focused on
interactive plant-based antimicrobial studies and to propose various methods and approaches
which could be considered when embarking on research in this field.
Experimental Approaches
Experimental Approaches
The terminology defining the possible interactions that may occur are often subject
to debate and interpretation [5], [9], [10], [11], [12], [13], thus for the sake of this review, the associated terminology should be defined.
The word “synergy” is derived from the Greek word “syn-ergo” meaning working together, and the resulting effect may be defined as a combination
that is significantly greater than the sum of its parts. Synonyms used include “polyvalent
activity” and “potentiation”. An “additive” or “summative” effect occurs when substances
added together will improve or increase efficacy. A “noninteractive”, “indifferent”
effect, or “zero interaction” reflects an expected linear response when two agents
are combined and show neither an additive nor antagonistic effect. “Antagonism” is
a phenomenon where two or more agents in combination have an overall effect which
is less than the sum of their individual effects [12], [14], [15]. For simplification, the terms “synergism, additive, indifferent, and antagonism”
will be used to describe the types of interactions.
There are a number of different methodologies that have been proposed to express antimicrobial
interactions. Many of these methods such as Etests, time-kill, and checkerboard methods
have been comparatively evaluated [16], [17], [18], [19]. Congruency in results for the evaluation of antibiotic synergy against Acinetobacter baumannii obtained between the three methods (Etests, time-kill, and checkerboard) varied between
51–72 % [18]. Comparative results generated from the time-kill and checkerboard method presented
only a 51 % value of congruency. Lewis et al. [19] favoured the Etest where antimicrobials of fixed concentrations are impregnated
on commercially available filter strips. This method, however, is not applicable for
plant-based antimicrobial studies in which the test antimicrobial is not a commercially
available sample at standard concentrations but an experimental plant sample whose
preparation is dependent on the undergoing study. Various authors have expressed concern
over the methods used to interpret synergy [5], [11], [12], [13], [20], [21], [22], [23]. In an editorial published in the Journal of Antimicrobial Chemotherapy [10], the interpretation of interactive methods was debated and a more conservative analysis
of synergistic interpretations encouraged (see section “The fractional inhibitory
concentration index” for further discussions on this). Mathematical models and statistical
approaches to validate antimicrobial interactions have been developed to allow for
a more reliable and quantitative assessment of pharmacological interactions [24], [25], [26], [27].
Considering plant-based antimicrobial studies, the use of different methodologies
range from the most basic disc diffusion assays found in earlier ethnobotanical studies
[28], [29], to more recent studies incorporating the sum of the fractional inhibitory concentration
index (ΣFIC) [30], [31], [32], [33], time-kill methods [34], [35], [36], [37], [38], and isobologram studies [3], [39], [40], [41], [42], [43].
“Basic” combination studies
The simplest form of determining synergy is by means of diffusion assays. Each independent
test sample (A or B) is placed in a well or on a disc. The combination (A + B) is
placed on a separate disc and the inhibition zone of the combination comparatively
examined with the independent test samples. Should the inhibition zone be larger in
A + B than either A or B then synergistic interactions are noted. Should the inhibition
zone be smaller in A + B than A or B independently, then antagonistic interactions
are noted. Although simple, these assays are subject to many variables which may influence
the results and should at the most be used as a qualitative guide only [40], [44], [45].
Basic minimum inhibitory concentration (MIC) assays may also be used to determine
interactions. The microdilution method is undertaken, and combinations are comparatively
assessed by incorporating the inhibitors at selected concentrations and combinations
[46], [47]. This arrangement of combinations formed by multiple dilutions is referred to as
the checkerboard method.
Some combination studies have incorporated impedimetric methods extrapolating synergy
by comparing growth as determined by optical density readings of single entities and
comparing these growth rates with that found when exposed to test substances in combination
[48], [49]. One drawback in using such a method is that the assessment of viability is not
always accurate when relying on turbidometric readings.
The sum of the fractional inhibitory concentration index (ΣFIC)
An algebraic equation to determine synergy by means of the ΣFIC is a widely accepted
means of measuring interaction. The ΣFIC is expressed as the interaction of two agents
where the concentration of each test agent in combination is expressed as a fraction
of the concentration that would produce the same effect when used independently [50]. The ΣFIC is then calculated for each test sample independently as specified in
the following equations:
The sum of the FIC or FIC index is thus calculated as: ΣFICI = FIC (*i) + FIC (*ii).
This basic equation has remained constant since inception. However, the interpretation
has evolved and varies from author to author. The interpretation of the ΣFIC index
as a numerical value is arbitrary, and the thresholds presented have stemmed from
the need to critically assess interactions that are clinically significant [13]. The earlier interpretations by Berenbaum [50] were very broad taking into account synergistic interactions having ΣFIC values
below one, antagonistic interactions above one, and additive interactions narrowly
focused on one. These interpretations make it easy to analyse isobolograms and have
been used in many papers describing interactions. A more conservative approach in
describing interactions was recommended by Odds [10], with interpretations described as synergistic (ΣFIC ≤ 0.5), antagonistic (ΣFIC > 4.0),
and noninteractive (ΣFIC > 0.5–4.0). The conservative approach takes into account
inherent variations when performing MIC doubling dilution assays. Unfortunately, the
“no interaction” range is a very broad one and makes no allowance for additive interpretations.
A number of recent reputable studies, mainly reported in ISI antimicrobial journals,
have incorporated an “additive” range into the interpretation for better clarification
of the data set [4], [5], [13], [19], [42], [51], [52], [53], [54], [55]. In the critical review by Bell [13], the need to include a broader range to interpret pharmacological interactions was
emphasised. Taking this into account, an additive range should be included, and it
is suggested that the interpretation of either synergistic (ΣFIC ≤ 0.5), additive
(ΣFIC > 0.5–1.0), noninteractive ΣFIC (> 1.0–≤ 4.0), or antagonistic (ΣFIC > 4.0)
should be used when describing in vitro antimicrobial interactions. A summary of the interpretative values given for the
ΣFIC in accordance with the corresponding authors is given in [Table 1]. Irrespective of the variations in interpretation, most authors are in agreement
that synergistic interactions should be considered only for ΣFIC values 0.5 and lower.
Table 1 Classification of the ΣFIC in accordance with corresponding authors.
|
Interaction
|
References
|
|
Synergy
|
Additive
|
Indifference
|
Antagonism
|
|
≤ 0.7
|
*
|
*
|
≥ 1.3
|
[133]
|
|
< 1.0
|
1.0
|
*
|
> 1.0
|
[50], [134]
|
|
≤ 0.5
|
> 0.5–4.0
|
> 4.0
|
[135]
|
|
≤ 0.5
|
1.0
|
*
|
≥ 2.0
|
[4]
|
|
≤ 0.5
|
> 0.5–1.0
|
> 1.0–< 2.0
|
≥ 2.0
|
[14]
|
|
≤ 0.5
|
*
|
> 0.5–≤ 4.0
|
> 4.0
|
[23]
|
|
≤ 0.5
|
0.5–< 1.0
|
≥ 1.0–< 4.0
|
≥ 4.0
|
[19]
|
|
≤ 0.5
|
*
|
> 0.5–4.0
|
> 4.0
|
[136]
|
|
≤ 0.5
|
*
|
> 0.5–4.0
|
> 4.0
|
[10]
|
|
< 0.5
|
0.5–1.0
|
≥ 1.0–4.0
|
> 4.0
|
[53]
|
|
< 0.5
|
0.5–≤ 1.0
|
> 1.1–≤ 4.0
|
> 4.0
|
[106]
|
|
≤ 0.5
|
*
|
> 0.5–< 4.0
|
≥ 4.0
|
[137]
|
|
≤ 0.5
|
> 0.5–0.75
|
0.76–2.0
|
≥ 2.0
|
[138]
|
|
≤ 0.5
|
> 0.5–1.0
|
> 1.0–≤ 4.0
|
> 4.0
|
[42], van Vuuren and Viljoen (recommended herein)
|
|
* Not given by author
|
Although using ΣFIC calculations to determine interaction appears to be the simplest
method, one needs to consider the limitations. The FIC method is based on the assumption
that half the concentration will provide half the effect. However, this is not always
the case, as two inhibitors may not always have identical dose responses [11]. The use of isobolograms which take into account combinations at various concentrations
provide a more realistic means of measurement.
The isobole method
The isobole method of determining interaction is possibly one of the oldest methods
used to express interactions, dating back to publications from 1870. Although well
established, negative connotations have been associated with this method proving it
unfavourable, until more recently when mathematical equations have been proposed to
validate the results [25], [26]. It is presently the favoured method for interactive assessment [13], [15]. The principle is based on the fact that interactions may vary depending on the
ratio in which the two inhibitors are combined. Although complicated, this method
gives a more accurate assessment of each agent when studied in various combinations.
The procedure involves the combination of two samples at various ratios. The MIC value
for each sample is determined independently and comparatively assessed against the
MIC value obtained in the ratio combination. This is expressed as a dose ratio response
on an isobole graph. The adjoining line of the two axes indicates the individual doses
and the isobologram can be interpreted by examining the data points of the ratios.
The classical interpretation of the isobole is where the data points fall below the
1 : 1 line; synergy is expressed [50]. Antagonism is noted for data points falling above the 1 : 1 line, and an additive
response is given when ratio points fall in the vicinity closest to or on the line.
To standardise interactive values with the more conservative approach recommended
by Odds [10], two additional lines are proposed, at the 0.5 : 0.5 and 4.0 : 4.0 axes ([Fig. 1]). These proposed additions will allow cross comparison between ΣFIC methods and
isobole interpretations.
Fig. 1 Isobologram which, if ratio points for two combined inhibitors fall in quadrants
A depicts synergy; B an additive effect; C a non-interactive effect; and D an antagonistic interaction.
Death kinetic (time-kill) assays
Time-kill studies provide descriptive information on the relationship between bactericidal
activity and the concentration of test substance [27]. Even though the methodology is labour intensive and requires a number of steps
where variables may be introduced, valuable information is given of the death kinetics
over time. The time-kill method has been praised as one of the best methodologies
to study synergy [48], even though earlier shunned by Berenbaum [56]. Further validation of the death kinetic method to assess synergy was given [57], and the method was commended from a clinical perspective. In an overview of the
various methods to test antimicrobial synergy with conventional antibiotics, the time-kill
method was found to be one of the most frequently employed, showed better sensitivities
and greater reproducibility [19]. Briefly, the principle involves exposing the inhibitor to a selected pathogen and,
at selected time intervals, aliquots are sampled and serially diluted. The dilutions
are plated out, incubated at optimum conditions for the test organism, and the colony
forming units (CFU) are counted and plotted logarithmically against time. Depending
on the curve of the dose response, either an additive, synergistic, or antagonistic
effect is noted ([Fig. 2]). Antagonism in time-kill methods may be defined as at least a100-fold increase
in colony counts whereas synergism, a 100-fold decrease in colony counts [48].
Fig. 2 Time-kill method of interpreting interactions when two samples are combined.
In spite of the positive recommendation of this method to describe antimicrobial interactions,
the method is not frequently used in plant-based studies. This is possibly due to
the labourious nature of repetitive dilution sampling. The positive aspect of this
method lies in the possibility to present a direct relationship in exposure of plant
test material to a pathogen. A cidal effect is monitored over time which is not possible
with the frequently used MIC assays.
The Various Levels at Which Antimicrobial Interactions May Be Explored
The Various Levels at Which Antimicrobial Interactions May Be Explored
Interaction between molecules
When examining the published literature and searching for scientific articles documenting
the interactions between molecules, findings were predominantly focused on essential
oil constituents. A quick review of some of our own studies demonstrates varied essential
oil compositions ranging anywhere from 25 to over 173 compounds in any given plant
[40]. It is thus not surprising that any of these compounds may interact to either enhance
or reduce pharmacological effects. With the sophisticated gas chromatography coupled
to mass spectrometry (GC-MS) and the multidimensional gas chromatography techniques
available today, detection of any number of compounds in a given plant may be undertaken.
Investigation of these interactions has thus become a more viable option than isolating
compounds from extracts and investigating interactions. Another limitation of isolating
compounds and investigating their interactive properties is that yields are usually
insufficient. With many essential oil studies, the identified compounds are done using
commercially available databases and retention indices.
In a review on synergism by Harris [58], the author documents on a number of earlier antimicrobial studies (between 1974–1996)
on volatile constituents that demonstrate synergistic interactions between constituents,
synthetic substances, and even ingredients within a formulation. Pattnaik et al. [59] noted that MICs from essential oils were in many cases lower than the major constituents
independently, suggesting that synergy between constituents may be contributing to
the enhanced activity. In another study, linalool was combined with methyl charvicol
at v/v ratios of 1 : 0; 0.8 : 0.2; 0.6 : 0.4; 0.4 : 0.6; 0.2 : 0.8, and 0 : 1 [60]. It was observed that when these two monoterpene alcohols are combined, a higher
efficacy is achieved, compared to when they are assayed independently. In another
study, the interaction of the major essential oil constituents of four Thymus species was examined by the MIC checkerboard method [61]. Various interactions ranging from indifferent to synergistic were observed when
combining carvacrol, thymol, 1,8-cineole, and p-cymene. No antagonism was noted, and the greatest synergistic interaction was observed
with the thymol : 1,8-cineole and thymol : p-cymene combination, having ΣFIC values of 0.125. In view of the international concern
on the use of antibiotic growth promoters in animal feeds, it was interesting to note
the application of combined essential oil constituents in controlling the antimicrobial
populations in the pig gut. The combination of carvacrol and thymol demonstrated synergism,
and recommendations for appropriate ratio studies to determine optimum synergistic
effects were recommended [62].
In-depth isobologram interpretations have been undertaken on essential oil constituent
interactions. Varied interactions were noted in a study where the pharmacological
interactions of a number of essential oil constituents were investigated [63]. Synergism was observed between (+)-β-pinene and carvacrol as well as between γ-terpinene and geranyl acetate when tested against Staphylococcus aureus. (+)-β-Pinene and (−)-menthone showed antagonism (ΣFIC value of 9.8), but interactions of
(+)-β-pinene with 1,8-cineole demonstrated synergy (ΣFIC value of 0.4), when tested together
against Candida albicans. The combination of trans-geraniol and E- and Z-(±)-nerolidol demonstrated an additive interaction against Bacillus cereus. For eugenol and E- and Z-(±)-nerolidol, an indifferent interaction against Escherichia coli was noted. These results demonstrate varied interactions and not only synergism.
All of these studies have one thing in common; they have focused on random essential
oil constituent combinations.
Although there have been a number of papers that have focused on structure activity
related antimicrobial studies of compounds within a plant [64], [65], [66], [67], very little research has been conducted on how the co-occurrence of combined compounds
contribute to efficacy. It has been questioned that independent activity related to
one or two specific constituents is questionable and that synergistic functions between
molecules are more probable [68]. In a study by Radulović [69], the major compound (68.6 % salicylaldehyde) from Filipendula vulgaris was isolated and found to be less active than the whole essential oil. When combined
in a 60 : 40 ratio with linalool (1.8 % composition in F. vulgaris oil), strong synergistic activity was noted. Interestingly, when salicylaldehyde
was combined with another essential oil component, methyl salicylate (2.4 %) in a
60 : 40 ratio, antagonism was observed. We too have noted that synergistic interactions
between molecules within a plant are evident. The two major essential oil components
from Osmitopsis asteriscoides identified by GC-MS were (−)-camphor (12 %) and 1,8-cineole (60 %) representing 72 %
accumulatively. Time-kill studies were performed on the pathogen C. albicans, where (−)-camphor demonstrated negligible antimicrobial activity and 1,8-cineole
indicated a cidal effect after 240 min. When these two major compounds were tested
in combination, a synergistic effect was noted having a cidal effect at 15 min [35]. Prediction that synergistic interactions occur only between major constituents
may not always be accurate. Earlier studies demonstrated that less abundant components
may interact synergistically [70]. This has been noted in other studies where the β-triketone complex of manuka oil was found to have poor bactericidal properties [34]. Similarly, we found this to be evident when investigating the antimicrobial activity
of the major constituents of Artemisia afra. The four major compounds (artemisia ketone, 1,8-cineole, α and β-thujone, which accounts for 51.9 % of the total composition) were investigated independently
and in various permutations. Results showed minimal antimicrobial activity against
Klebsiella pneumoniae. It was thus postulated that the minor compounds either independently or in combination
contribute to the antimicrobial activity [37]. As noted in these interactions, when examining whole essential oils, predictions
are complex and not only should major or minor compounds be considered, but one also
needs to consider the stereochemistry of compounds. While it is known that biological
activity is influenced by the enantiomeric configuration, the overall antimicrobial
activity of different enantiomers may be additionally affected by interaction with
other compounds. To demonstrate this, a study was undertaken on the different enantiomers
of limonene in combination with 1,8-cineole. Isobologram plots for S. aureus demonstrated similar antagonistic activity when exposed to the combinations of 1,8-cineole
with (+) and (±)-limonene. However, with (−)-limonene, synergism was evident at selected
ratios. Differences in activity were clearly noted with the other two pathogens studied
(Pseudomonas aeruginosa and Cryptococcus neoformans), thus highlighting the significance of stereochemistry in antimicrobial combination
studies.
Compound interaction studies on nonessential oil components include the combined effects
of cinnamaldehyde with catechin, quercetin, or eugenol, tested against wood decay
fungi with the aim to provide a rational in natural wood preservation. Varied interactions
were noted ranging from synergistic to antagonistic when tested against the two fungal
test organisms Lenzites betulina and Laetiporus sulphureus [71]. Another example where poorer antimicrobial activity is noted for the isolated compound
rather than the combined compounds was reported in a study of linoleic and oleic acid,
isolated from Helichrysum pedunculum, which were found to have higher activity against S. aureus and Micrococcus kristinae in combination (MIC value 0.05) than independently (MIC values 1.00) [72]. More recently [73], it was demonstrated that the biological effects observed for the major compounds
of Ocimum gratissimum were not responsible for the overall effect of the essential oil. These types of
studies reinforce the concept of a multi-targeted approach in therapeutic strategies
and prove the hypothesis formulated by Tyler [74], that searching for potent antimicrobial compounds is becoming more and more improbable
and that research should be moving towards the investigation of a combination of substances
to achieve efficacy.
Interactions between different plant parts or fractions
Although it is evident that the many constituents within plants interact, it should
also be noted that other interactions may occur between groups of molecules or fractions
of the plant. For many species there is a strong distinction in the chemistry between
the subterranean and above ground plant organs. If, for example, roots and leaves
are combined, then the number of “active” compounds may be increased, and possibly
an increased chance of synergistic interactions may occur. This may be why there are
numerous anecdotal reports of plants used in combination therapy, i.e., when plant
parts such as roots and leaves are combined and used in therapeutic regimens [75], [76], [77], [78], [79]. Even though the ethnobotanical use of many plants incorporates mixes of the different
plant parts, there has been very little scientific evidence to support such interactive
efficacies. There have been numerous screening studies that have investigated the
antimicrobial activity of different plant parts such as fruit, leaf, root, barks,
seeds, etc.; however, these have been investigated separately and not in combination
[80], [81], [82], [83], [84], [85]. Our studies on the various plant parts of Croton gratissimus was undertaken on the ethnopharmacological basis that these various parts are often
used in combination. The root, leaf, and bark extracts were investigated singularly
and combined in various ratios to establish possible interaction. The MIC value (0.4 mg/mL),
ΣFIC (0.4), and isobologram results (all ratios depicting synergy) for C. neoformans validate the traditional use of a root: leaf combination [41]. Up until recently, no other studies, other than our own work, could be found where
different parts of the same plant are combined and investigated for antimicrobial
efficacy [41], [86], [87]. It is encouraging to see such studies now being published, in which the individual
and combined phenolics within the Olea europaea plant extract have been studied [88]. The results indicated that the combined phenolics had significantly higher antimicrobial
activity than the individual phenolics investigated within the plant. Only one earlier
study could be found that partially addresses the interactions that may occur between
plant fractions. Garlic oil and allyl alcohol, both derived from Allium sativum, were combined and their interaction evaluated against Candida utilis. Isobolograms were used to interpret synergistic ΣFIC values of between 0.37–0.42
[89].
Aromatic plants have an additional component of chemical complexity— the volatile
constituents. In healing rituals, the volatiles may be administered selectively (inhalation)
or the volatiles and nonvolatiles may by applied collectively, e.g., a poultice placed
directly onto a wound or the alcoholic extraction (tincture) of crude plant material.
It has been well recorded that extracts of aromatic plants have superior activity
over the essential oils [41], [90], [91]. In our own studies, we explored the possible interaction between the volatile and
nonvolatile fractions to yield greater antimicrobial activity. To test this hypothesis,
a number of plants were investigated, i.e., three Pelargonium species (P. graveolens, P. quercifolium, P. tomentosum) [92], Plectranthus grandidentatus [40], three Salvia species (S. africana-caerulea, S. africana-lutea, and S. lanceolata) [86], and Tarchonanthus camphoratus [87]. These studies examined the need for coexistence of volatile and nonvolatile constituents
to enhance antimicrobial efficacy. It is widely accepted that the administration of
an infused oil may act as a penetrative enhancer [1], and possibly the synergistic interactions noted may be a result of improved solubility
and bioactivity of the active principles.
Interactions between different plant species
Currently available on the market are phytomedicines which are sold as whole extract
combinations, for example, Gingko biloba with Echinacea. It is believed that synergistic interactions are responsible for their therapeutic
efficacy [15]. Many traditional healing practises prescribe plant combinations from different
species to treat diseases. A comprehensive study has been undertaken on the ethnobotanical
use of plant mixtures, in which 170 plant species from Cuba were examined for their
combined medicinal use. Sixty-one combinations were attributed to anti-infective applications
[93]. In African traditional medicine it is well known that traditional healers often
combine various plant species in order to enhance efficacy. A number of instances
where plants have been combined for the treatment of microbe related infections have
been found in the ethnobotanical literature. Some documented accounts include the
combination of Portulaca quadrifida with Monadenium lugardiae to treat stomach complaints, Trichilia emetica with Cyathula natalensis for leprosy, and Momoridica foetida with Pittosporum viridiflorum for boils. Various combinations of C. gratissimus have been used. Accounts of the administration with other species have been noted,
e.g., for the treatment of swellings, the bark of C. gratissimus is combined with the root of Amaryllidaceae species and rubbed into incisions. Also
noted is the use of the bark of C. gratissimus and Ocotea bullata in combination, which are powdered and blown into the womb to treat uterine disorders
[76].
Given the fact that it is common traditional practice to combine medicinal plants,
it was surprising to find so little published on plant to plant interactions. Previous
studies include a study on the combination of Thymus vulgaris with Pimpinella anisum, two plants combined in Iraqi folk medicine. Both essential oils and methanol extracts
were studied against nine test organisms, and predominantly additive interactions
were noted [94]. Tea tree (Melaleuca alternifolia) and lavender (Lavandula angustifolia) essential oils were combined and tested against the dermatophytes Trichophytum rubrum and Trichophytum mentagrophytes var. interdigitale. Various combinations were prepared, and results presented in isobolograms, demonstrating
an antimycotic effect [95].
Artemisia afra is one of the oldest and most widely used plants in African traditional medicine
[96], [97]. It is commonly used to treat respiratory infections such as coughs, colds, lung
inflammation and often combined with plants such as Lippia javanica, Agathosma betulina, Osmitopsis asteriscoides, Eucalyptus globulus,
Zanthoxylum capense, Leonotis microphylla, Tetradenia riparia, and Allium sativum [75], [76], [98]. In spite of these numerous reports in the ethnobotanical literature, very little
research has been dedicated to validate these combinations. In our own combination
studies on anti-infective African traditional medicines, A. afra was combined with L. javanica. The objective was to scientifically validate the concomitant use of these two coveted
ethnomedicinals to treat respiratory infections. A time-kill assay was undertaken
against the respiratory pathogen K. pneumoniae. Essential oil obtained from L. javanica (0.25 %) and A. afra (0.25 %) were run independently and in combination (L. javanica and A. afra together totalling 0.25 %). Artemisia afra when studied independently showed initial microbial destruction within one hour,
but regrowth after 24 h. For the L. javanica oil at 0.25 %, death kinetics was observed within 40 min, but regrowth after four
hours. When the two plants were combined, a bactericidal effect was maintained for
the full 48 hours of testing. This synergistic effect scientifically validates the
combined use of L. javanica and A. afra for the treatment of respiratory infections associated with K. pneumoniae and corroborates the traditional use of these two plants when administered in combination
[40]. More recent results of the combination of the essential oils of A. afra with three medicinal aromatic plants, Agathosma betulina, Eucalyptus globulus, and O. asteriscoides, displayed predominantly additive interactions [99]. In an earlier combination study, we investigated the ethnobotanical use of Salvia chamelaeagnea in combination with Leonotis leonurus to treat respiratory infections. Individual extracts and a combination of the aerial
parts of S. chamelaeagnea and L. leonurus were evaluated for the in vitro antibacterial activity (ΣFIC index presented as data points in isobolograms). When
the two extracts were combined, synergistic actions were observed for the Gram-positive
bacteria while antagonism, synergism, and/or additive actions were observed for the
various ratios tested on studies with the Gram-negative bacteria [39].
Plant combinations with the potential to increase preservative efficacy in foods have
recently been given more attention. The trend toward a more natural and greener approach
to consumerism together with the economic benefit that may concur make this area of
research attractive. Thus, a number of studies have been undertaken on food systems
in the hope to achieve better antimicrobial effects in combination. A disc diffusion
study was undertaken where Cinnamomum cassia was combined with Allium tuberosum and the fruit of Cornus officinalis in a triple combination at varying ratios i.e., 1 : 1 : 1, 8 : 1 : 1, 6 : 6 : 1,
1 : 6 : 6, 1 : 8 : 1, and 3 : 1 : 2. The combination 8 : 1 : 1 (C. officinalis: C. cassia: A. tuberosum) was found to possess antimicrobial efficacy against a wide range of test organisms.
When applied to food systems, the combination retained antimicrobial activity [100]. Another interactive study with effective food preservation as an outcome was undertaken
using isobolograms to interpret interactions. Fractions of Eucalyptus dives and Coriandrum sativum were combined and investigated against 12 test organisms. Of all the test organisms,
Yersinia enterocolitica was the most susceptible to synergism. Other interactions varied between additive
to antagonistic [101]. More recently, Origanum vulgare and Thymus vulgaris were combined and found to have an additive effect against food spoilage bacteria
[102]. In another study, the interactive combination of cranberry, blueberry juice, and
grape seed extract was antimicrobially tested against Helicobacter pylori. This study was undertaken on the assumption that a diet rich in phytocompounds may
act prophylactically to ward off infection. Of the five different combinations formulated,
the permutation having cranberry juice extract (75 %) with blueberry juice extract
(10 %) and grape seed extract (15 %) demonstrated the highest synergy [103]. Multiple combination studies are challenging due to the endless number of permutations
which exist to produce complex formulation. The development of predictive software
using factorial designs to optimize experimental design may offer a solution to simplify
the complexity.
A study on the combined effect of Origanum vulgare and Vaccinium macrocarpon was undertaken using the disc diffusion assay. Antimicrobial activity against Vibrio parahaemolyticus was best noted in a 1 : 1 combination [36]. Rosemary and clove essential oils have been combined and the antimicrobial efficacy
reported using both MIC methods and time-kill studies. This comprehensive study was
undertaken on a number of different pathogens and various MIC ratios depicted mostly
additive effects against the test bacteria; Staphylococcus epidermidis, S. aureus, Bacillus subtilis, E. coli, Proteus vulgaris, and P. aeruginosa. The fungal test organisms, however, demonstrated either synergy (C. albicans) or antagonism (A. niger). The time-kill studies reported that combinations in lower concentrations were not
sufficient to produce a cidal effect. Only concentrations twice that of the MIC value
had a lethal effect [104]. Using synergistic principles, some plants were evaluated for the prevention of
Cassava root rot during storage. Garlic, Landolphia owerrience, and Garcinia kola were investigated independently and in various 1 : 1 combinations. The combination
of garlic with G. kola demonstrated the highest inhibition preventing rot during 14 days of storage [105]. Essential oil combinations of oregano (Origanum vulgare) with basil (Ocimum basilicum), lemon balm (Melissa officinalis), marjoram (Origanum majorana), rosemary (Rosmarinus officinalis), sage (Salvia triloba), and thyme (T. vulgaris) have been investigated against Bacillus cereus, Escherichia coli, Listeria monocytogenes, and Pseudomonas aeruginosa. All interactions either demonstrated an additive or indifferent effect in the MIC
checkerboard method [106]. Using time-kill methods, a study was undertaken on Melaleuca alternifolia oil which was blended with a polar fraction of manuka (Leptospermum scoparium). Death kinetics demonstrated synergistic interactions [34]. Interactions between the isolated compound polygodial and plant species such as
Perilla frutescens and Licaria puchuri-major have yielded varied results depending on the pathogen studied [107], [108]. Another study focusing on selected unrelated compounds and the combination thereof
was undertaken whereby the antimicrobial action of Staphylococcus aureus produced a synergistic effect when berberine, a common alkaloid found in a variety
of plant species, was combined with 5′-methoxyhydnocarpin [109].
After reviewing the literature and examining the scientific interactive antimicrobial
studies presented, it is clear that reports on antagonistic interactions seem to be
largely ignored or possibly rejected by phytotherapy journals. It is thus encouraging
to see that a study that focused on the antagonistic effects of two herbal extracts
(Rhizoma Coptidis and Fructus Evodiae) in a traditional Chinese medicine formula have
recently been published. Findings suggest that the two samples have opposing effects
[110].
Antimicrobial plant interactions with nonbotanical antimicrobial agents
Of all the interactions studied, research into phytoconstituents in combination with
nonbotanical chemical entities has been the most widely studied. Studies range from
plant interactions with preservatives to interactions with conventional antimicrobials.
[Tables 2] and [3] record these studies including a summary with the salient outcomes achieved. Interactive
interpretation is given according to the respective authors. For many of the plants
whose allopathic combination studies were reviewed, either an additive or synergistic
effect is presented. Very few reports have documented antagonism despite the fact
that it is not unusual to encounter adverse drug reactions with herbal medicines [15]. In our own studies we assessed the interaction between a selection of popular commercial
oils (Melaleuca alternifolia, Thymus vulgaris, Mentha piperita, and Rosmarinus officinalis) and conventional antimicrobials (ciprofloxacin and amphotericin B). The initial
objective was to determine if a synergistic pattern predominates as noted in other
similar studies reported in literature. Whilst some synergistic and additive interactions
were evident between essential oils and antimicrobials, antagonistic interactions
were also highlighted. It was interesting to note that when Melaleuca alternifolia (tea tree) oil which is often recommended for treatment of skin ailments, was combined
with ciprofloxacin and tested against Staphylococcus aureus, antagonism was noted for all ratios in the isobologram [111]. This study highlights that caution should be adhered to when combining natural
products with allopathic antimicrobials and addresses the proposal by Cuzzolin et
al. [112], that there is a need for more systematic interactive studies to be undertaken to
identify unfavourable combinations.
Table 2 The combination of plants with conventional antibiotics.
|
Plant derived test substance
|
Non-plant derived test substance
|
Test organism
|
Interaction
|
References
|
|
Santolina chamaecyparissus
|
clotrimazole
|
Candida albicans
|
synergistic when comparing MIC data
|
[139]
|
|
Agastache rugosa and major compound estragole
|
ketoconazole
|
Blastoschizomyces capitatus
|
isobologram depicting synergy
|
[140]
|
|
Bidwillon isolated from Erythrina variegata
|
mupirocin
|
Staphylococcus aureus
|
FIC values range between 0.5–1
|
[30]
|
|
Pomegranate extract
|
chloramphenicol
|
Staphylococcus aureus
|
FIC values range between 0.03–1
|
[31]
|
|
gentamicin
|
FIC values range between 0.13–4
|
|
ampicillin
|
FIC values range between 0.03–1
|
|
tetracycline
|
FIC values range between 0.03–1
|
|
oxacillin
|
FIC values range between 0.03–1
|
|
Mentha piperita essential oil and menthol
|
ampicillin
|
Escherichia coli
|
FIC values range between 1–2
|
[53]
|
|
erythromycin
|
FIC values range between 1–2
|
|
gentamicin
|
FIC values range between 1–1.25
|
|
oxytetracycline
|
FIC values of 0.5
|
|
Kola nitida
|
ciprofloxacin
|
Escherichia coli
|
potentiation for all antibiotics tested
|
[141]
|
|
pefloxacin
|
|
levofloxacin
|
|
Cassia fistula fruit solution
|
amoxycillin
|
Salmonella enterica (48 isolates)
|
FIC method indicated synergism for 80 % strains tested; no antagonism noted
|
[142]
|
|
Essential oils from Cedrus atlantica, Styrax tonkinensis, Juniperus communis, Lavandula angustifolia,
Melaleuca alternifolia, Pelargonium graveolens, Pogestemon patchouli, and Rosmarinus officinalis
|
ketoconazole
|
Aspergillus niger
|
FIC indices ranging from 0.52–1
|
[143]
|
|
amphotericin B
|
Aspergillus flavus
|
|
Pelargonium graveolens and main constituents citronellol, geraniol, triacetin
|
norfloxacin
|
Bacillus subtilis
|
FIC indice 0.5
|
[144]
|
|
Bacillus cereus
|
FIC indices 0.5; synergy in isobole method for plant oil and norfloxacin
|
|
Staphylococcus aureus (2 strains)
|
FIC indices ranging from 0.37–0.5; synergy in isobole method for plant oil and norfloxacin
|
|
Escherichia coli
|
FIC indice 0.57
|
|
Allium species (essential oils)
|
ketoconazole
|
Trichophytum spp.
|
FIC indices ranging from 0.09–0.75
|
[145]
|
|
α-Mangostin isolated from Garcinia mangostana
|
ampicillin
|
Enterococcus faecalis (8 strains)
|
FIC values range between 0.5–1
|
[146]
|
|
gentamicin
|
Staphylococcus aureus (9 strains)
|
|
Catha edulis
|
tetracycline
|
Streptococcus oralis
|
4-fold potentiation
|
[147]
|
|
tetracycline
|
Streptococcus sanguis
|
2-fold potentiation
|
|
penicillin G
|
Fusobacterium nucleatum
|
4-fold potentiation
|
|
Eight Chinese medicinal plants
|
penicillin G
|
resistant and standard strains of Staphylococcus aureus
|
% inhibition of combination varies between < 1 (synergistic) to 75.4
|
[148]
|
|
gentamicin
|
% inhibition of combination varies between < 1 (synergistic) to 104.5
|
|
ciprofloxacin
|
% inhibition of combination varies between < 1 (synergistic) to 107.3
|
|
ceftriaxone
|
% inhibition of combination varies between < 1 (synergistic) to 71.5
|
|
Sophoraflavanone G isolated from Sophora flavescens
|
gentamicin
|
11 strains of oral bacteria
|
FIC values range between 0.28–0.75
|
[149]
|
|
15 traditional Indian plants
|
tetracycline
ciprofloxacin
|
Staphylococcus aureus and Escherichia coli
|
synergism or “neutralism” when investigating inhibition zones
|
[150]
|
|
Melaleuca alternifolia, Origanum vulgare, and Pelargonium graveolens
|
amphotericin B
|
5 different Candida strains
|
isobolograms demonstrating P. graveolens oil with amphotericin B as the most synergistic combination
|
[151]
|
|
Galangin isolated from Alpinia officinarum
|
gentamycin
|
Staphylococcus aureus
|
FIC values range between 0.18–0.255
|
[32]
|
|
Thymus eigii
|
vancomycin and erytromycin
|
13 test organisms
|
antagonism determined by zone inhibition
|
[152]
|
|
Rhus coriaria, Psidium guajava, Lawsonia inermis, Sacrpoterium spinosum
|
oxytetracyclin
|
Staphylococcus aureus
|
synergy determined by zone inhibition
|
[153]
|
|
gentamicin
|
synergy/antagonism determined by zone inhibition
|
|
enrofloxacin
|
antagonism determined by zone inhibition
|
|
sulphadimethoxin
|
synergy determined by zone inhibition
|
|
Melaleuca alternifolia, Thymus vulgaris, Mentha piperita, and Rosmarinus officinalis essential oils
|
ciprofloxacin and amphotericin
|
Staphylococcus aureus, Klebsiella pneumoniae, and Candida albicans
|
antagonism mainly noted with amphotericin B: essential oil combination
|
[111]
|
|
Croton zehntneri
|
gentamicin and tetracycline
|
Staphylococcus aureus and Pseudomonas aeruginosa
|
activity increased by 42.8 % against P. aeruginosa in combination
|
[154]
|
|
Thespesia populnea
|
oxytetracycline
|
12 bacterial strains
|
highest synergy noted for Shigella boydii using the disc diffusion method
|
[155]
|
|
Origanum vulgare, Pelargonium graveolens, and Melaleuca alternifolia
|
nystatin
|
5 different Candida strains
|
O. vulgare essential oil and nystatin indicate most prominent synergy with FIC indices between
0.11 and 0.17
|
[156]
|
|
Rhus coriaria, Sacropoterium spinosum, Rosa damascene
|
oxytetracycline, penicillin G, cephalexin, sulfadimethoxine and enrofloxacine
|
Pseudomonas aeruginosa (3 clinical strains)
|
synergy
|
[157]
|
|
Ocimum sanctum essential oil
|
fluconazole and ketoconazole
|
16 fluconazole-resistant Candida isolates
|
FIC values ranging from mostly synergistic (0.25–0.50) to indifferent (0.52–0.93)
|
[158]
|
|
Melaleuca alternifolia oil
|
tobramycin
|
Escherichia coli and Staphylococcus aureus
|
synergy demonstrated with time-kill methods
|
[159]
|
|
Punicalagin isolated from Punica granatum
|
fluconazole
|
Candida albicans
|
synergy demonstrated with disc diffusion, MIC, and time-kill methods
|
[160]
|
|
Eugenol, thymol, carvacrol, cinnamaldehyde, allyl, and isothiocyanate
|
tetracycline, ampicillin, penicillin G, erythromycin, bacitracin, and novobiocin
|
Salmonella typhimurium, Escherichia coli, Streptococcus pyogenes and Staphylococcus aureus
|
FIC indices mostly indicating synergy with strongest synergy noted (FIC 0.11) with
carvacrol: penicillin G combination against S. aureus
|
[161]
|
|
Myrtus communis essential oil
|
amphotericin B
|
Candida albicans and Aspergillus niger
|
FIC indices and isobologram indicate synergy
|
[162]
|
|
Thymus maroccanus and Thymus broussonetii
|
amphotericin B and fluconazol
|
Candida albicans
|
synergistic FIC indices ranging between 0.27–0.49
|
[163]
|
Table 3 The combination of plants with nonbotanical agents other than conventional antibiotics.
|
Plant derived test substance
|
Non-plant derived test substance
|
Test organism
|
Interaction
|
Reference
|
|
Origanum vulgare and Vaccinium macrocarpon
|
lactic acid
|
Vibrio parahaemolyticus
|
total time-kill inhibition throughout 10 h tested
|
[36]
|
|
Ocimum basilicum (anise variety)
|
5 % NaCl
|
Lactobacillus curvatus
|
synergy using indirect impedance method
|
[60]
|
|
Thymol
|
potassium sorbate
|
Escherichia coli
|
FIC values range between 0.5–1.1
|
[52]
|
|
Listeria innocua
|
FIC values range between 0.3–0.8
|
|
Salmonella typhimurium
|
FIC values range between 0.5–1.0
|
|
Staphylococcus aureus
|
FIC values range between 0.3–0.8
|
|
Carvacrol
|
Escherichia coli
|
FIC values range between 0.5–0.8
|
|
Listeria innocua
|
FIC values range between 0.4–0.8
|
|
Salmonella typhimurium
|
FIC values range between 0.4–0.6
|
|
Staphylococcus aureus
|
FIC values range between 0.4–0.9
|
|
Rosmarinus officinalis
|
butylated hydroxyanisole
|
Escherichia coli and Staphylococcus aureus
|
synergy for most points depicted on the isobolograms
|
[43]
|
|
Oregano & oregano : cranberry extracts
|
sodium lactate (1–2 %)
|
Listeria moncytogenes
|
total time-kill inhibition between days 10–15
|
[38]
|
|
Carvacrol, thymol, and eugenol
|
nisin
|
Listeria moncytogenes
|
optical density readings indicate reduction in growth
|
[123]
|
|
diglycerol fatty acid esters
|
varied interaction depending on compound studied
|
|
Melaleuca alternifolia, Leptospermum scoparium, and Leptospermum morrisonii
|
chlorhexidine digluconate
|
Streptococcus mutans and Lactobacillus plantarum
|
CFU reduction 4- to 10-fold depending on the combination
|
[126]
|
|
Thymol, carvacrol, citral, eugenol, geraniol
|
acetic, citric, lactic, and pyropolyphosphoric acids
|
Salmonella typhimurium
|
no synergistic effects noted
|
[164]
|
|
Oils of fennel, anise, and basil
|
benzoic acid and methyl-paraben
|
Listeria monocytogenes and Salmonella enteriditis
|
synergy detected in five of the 16 combinations studied
|
[165]
|
|
Thymus vulgaris, Rosmarinus officinalis, Origanum vulgare
|
lactic acid
|
Listeria monocytogenes
|
synergy at lower concentrations with highest synergistic activity noted for thyme : lactic
acid and rosemary : lactic acid combinations
|
[166]
|
|
Eucalyptus oil, tea tree oil, and thymol
|
chlorhexidine digluconate
|
Staphylococcus epidermidis
|
thymol: chlorhexidine digluconate combination demonstrated the most synergistic interactions
with FIC values of 0.25
|
[127]
|
|
Nine different plant essential oils
|
enterocin AS-48
|
Listeria monocytogenes
|
reduction in CFU for combinations
|
[167]
|
|
Origanum vulgare subsp. hirtum
|
nisin
|
Salmonella enteriditis
|
combination of O. vulgare essential oil at 0.9 % and nisin at 1 000 IU/g demonstrated the best synergistic
activity over time
|
[168]
|
Many of the methods employed to depict synergy between plants and nonbotanical components
other than conventional antibiotics ([Tables 2] and [3]) are based on the experimental methods described herein. However, some studies have
used alternate approaches to prove synergistic interactions. Such studies include
the investigation of synergistic effects between catechin, an extract of green tea
which was combined with ciprofloxacin using in vivo studies on a rat model. It was confirmed that the combination resulted in a statistically
significant decrease in bacterial growth [113]. Kurita and Koike [114] examined the combination of ethanol, sodium chloride, or acetic acid with 19 essential
oil components. Using an agar dilution method incorporating various combinations,
the interactions were analysed over a 20-day period. Studies were undertaken with
seven fungal species. Generally, a synergistic effect was noted when the variables
were combined in pairs, threes, or altogether. In another study, synergistic interactions
were also determined over a 21-day period against Penicillium notatum, however, with the incorporation of volatile compounds in an atmospheric jar. Synergistic
interactions were noted for six combinations (ethanol: carvacrol; sulphur dioxide:
carvacrol; sulphur dioxide: isothiocyanate; sulphur dioxide: cinnamaldehyde; isothiocyanate:
cinnamaldehyde; cinnamaldehyde: carvacrol) [115]. A recent study demonstrated synergistic interactions with carvacrol. These combinations
comprised of carvacrol with ciprofloxacin and carvacrol with amphotericin B against
Bacillus cereus and C. albicans, respectively. Additionally, eugenol with ciprofloxacin or amphotericin B was synergistic
when tested against E. coli and C. albicans, respectively [63].
An area that has been sorely neglected is the incorporation of botanicals within formulations
to achieve an enhanced naturally subsidised pharmaceutical product. Nostro et al.
[116] examined the synergistic interactions of Calamintha officinalis with EDTA in cream formulations. More recently, Artemisia afra, Eucalyptus globulus, and Melaleuca alternifolia were encapsulated into diastearoyl phosphatidylcholine and diastearoyl phosphatidylethanolamine
liposomes. The ΣFICs were calculated in order to determine if the incorporation of
essential oils would enhance the antimicrobial activity of the formulation. Synergistic
to additive interactions were noted for encapsulated E. globulus (ΣFIC values 0.25–0.45) and M. alternifolia (ΣFIC values 0.26–0.52) formulations [33].
Future Considerations
Future Considerations
The reductionist approach in studying natural products
For decades, phytochemists have been isolating natural products in the hope to find
an antimicrobially active molecule comparable in activity to the allopathic antimicrobials
available today. Yet, no single plant-derived antibacterial has been commercialised
[117]. Perhaps this reductionist approach which limits complexity and variability is somewhat
short-sighted when we consider the convolution of plants and the many compounds (major
and minor) that may contribute to the overall activity of the plant. Allopathic medicine
has realised the importance of studying the interaction between molecules for several
years. Typical titles would include papers such as “In vitro synergy studies based on tazobactam/piperacillin against clinical isolates of metallo-β-lactamase-producing Pseudomonas aeruginosa” [118] and “Synergic activity, for anaerobes, of trovafloxacin with clindamycin or metronidazole:
chequerboard and time-kill methods” [119]. [Fig. 3] shows the number of papers reporting on pharmacological interactions for three journals
(two plant based and one antimicrobial) over a ten-year period. The Journal of Antimicrobial
Chemotherapy has consistently published a number of papers on this subject over the
ten-year period (a total of 161). Ironically, the Journal of Ethnopharmacology and
Planta Medica only carried 61 and six papers, respectively, over the same period.
This is ironic for several reasons. The very nature of ethnopharmacology is based
on traditional healing practices where crude extracts are administered, not molecules,
and hence one cannot ignore the possible interactions between the various constituents
and different species. Furthermore, researchers working in the field of ethnopharmacology
often justify their projects based on traditional practices, but often the methodology
followed is divorced from the real-life application in a traditional setting. Although
it remains a rewarding and challenging exercise to search for the active principles
in complex crude mixtures, it is not surprising that so many papers following this
reductionist approach conclude that “the crude extract was more active than the isolated
molecules”.
Fig. 3 Number of papers reporting on pharmacological interactions for three journals (Journal
of Antimicrobial Chemotherapy, Journal of Ethnopharmacology, and Planta Medica) over
a ten-year period.
Natural products and antimicrobial mode of action
The mode of action of conventional antimicrobials, both independently and in combination
therapy, have been extensively studied. The mechanisms by which agents act on the
cell wall interfere with biological pathways, and other more complex interactions
have also been explored [57]. With respect to natural product combinations, the mechanism of synergy may be attributed
to complex multi-target effects, pharmacokinetic or physiochemical properties, neutralization
principles, or even therapeutic approaches [121]. Gilbert and Alves [120] have hypothesised on the efficacy of whole plant extracts rather than isolated molecules.
Furthermore, an extensive review on possible modes of action of natural products with
allopathic antimicrobials has been undertaken [122]. Targets include receptor site modification, enzymatic degradation, reduced accumulation
of drug within the bacterial cell, decreased membrane permeability, and efflux pumps.
Even though these possible modes of action have been addressed, supporting studies
to confirm these mechanisms are sorely lacking, especially with respect to combination
therapy. When reviewing the literature on phytotherapeutic combinations, the elucidation
of mode of action for both inhibitors are rarely reported in spite of the authors'
efforts to include this as recommendations for further study [52], [101], [123], [124]. In a study on the synergistic interaction of Punica granatum (methanol extract) with a range of antibiotics, the authors allude to the mode of
action whereby the extract plays a role in efflux inhibition enhancing the uptake
of a conventional drug [31]. In another study on the combination of an isoflavanone from Erythrina variegate with mupirocin, the mechanism of action is thought to involve bacterial cell membranes;
however, further studies are recommended for confirmation [30]. One notable study that focused on mode of action with respect to combined inhibitors
was the investigation of berberine with 5′-methoxyhydnocarpin where the mode of action
was attributed to the effect of 5′-methoxyhydnocarpin blocking the Nor A pump and
thus potentiating the antibiotic action of berberine [109]. This valuable insight into specific modes of action should be encouraged in future
endeavours, keeping in mind that exploring this area of research may be extremely
complex. One needs to consider the variability of possible interactions that may occur
within this phytochemical pool, not only within a single extract but in various combinations.
Furthermore, predominant mechanisms of action may be potentiated by other less effective
modes of action [7] and vice versa.
Biofilm inhibition from combined phytomedicinals
Plant-based antimicrobial studies on planktonic microorganisms have been given extensive
priority. The inhibition of biofilms, whether on independent or combined plant inhibitors,
however, has been largely neglected. Combination studies are sparse, and the only
interaction predominant in the literature is the combination of phytomedicinals with
chlorhexidine digluconate, a skin antiseptic commonly used in clinical settings. Studies
include the combined effect of Eucalyptus essential oil and the monoterpene 1,8-cineole with chlorhexidine digluconate, for
which mainly synergistic interactions were found against C. albicans, E. coli, P. aeruginosa, S. aureus (including a methicillin resistant strain) biofilms [125]. Previous studies have shown that a number of essential oils together with chlorhexidine
digluconate are effective in inhibiting biofilm cultures [126], [127].
In another biofilm combination study, two diterpenoids, salvipisone and aethiopinone,
isolated from the roots of Salvia sclarea and combined with beta-lactam antibiotics, demonstrated synergy. It was postulated
that the mechanism of action may be due to cell surface hydrophobicity or cell wall
permeability [128]. Studies such as this provide not only valuable information on biofilms but also
offer explanations on possible modes of action.
Conclusions
Conclusions
There has been a recent increase in awareness towards the concept of synergy within
phytomedicine, as noted in a number of review style articles [6], [15], [120], [121]. In conjunction with earlier publications on synergistic principles [9], [56], [122], [123], [129], [130], the validation of multiple phytotherapy has provided a much needed platform with
which to expand future research in this area. In particular, research into antimicrobial
combinations may yield new developments that may address the ever increasing concern
towards antimicrobial resistance. It has been shown that resistance to crude extracts
occurs less than resistance to single actives [131]. Thus, the search for single targeted molecules may not yield long-term solutions
in combating antimicrobial resistance. For plants to rely on a single compound in
their biochemical warfare with pathogens would be equivalent to relying on the “single
golden bullet” approach, and thus, as researchers investigating the activity of single
compounds, we would be ignoring the evolutionary approach that plants may have developed
various metabolic mechanisms for the production of structurally and functionally diverse
compounds to overcome emerging resistance. To ensure future success in natural product
research, we encourage interactive phytochemical studies with existing practices in
the hope that developments may be used as a foundation and driving force in the much
needed discovery of novel chemotherapeutic agents.
It is recommended that future development in the field of phytosynergy should consider:
-
The selection criteria for the two inhibitors. This should be clearly defined and
justification should be given for the choice of test substances to be examined in
combination.
-
Classification of synergistic interactions should be more conservatively evaluated
taking into account inherent doubling dilution variations noted in MIC methodology.
-
Even though it is popular to report synergistic interactions, antagonism should be
given the same priority.
-
Combinations involving more than two plant entities should be examined where applicable.
-
Combination studies involving biofilm inhibition should be considered.
-
Articles addressing the mechanism of action of synergistic interactions should be
given precedence. It is encouraging to note that some attention to this has been given
[6], [120]. The criteria addressed include receptor or site modification, enzymatic degradation,
accumulation of antibiotic within bacterial cell, decreased outer membrane permeability,
and efflux pumps. This area of research has been greatly neglected as many studies
just report on the interactions that occur.
With such validations in place, the justification and development of antimicrobial
combinations could lead to patentable entities making research in the field of phytosynergy
commercially relevant [132]. New techniques, such as metabolomics and the dual applications of chemometric data
analysis methods, are providing the researcher with new tools to explore this fascinating
phenomenon which will undoubtedly become increasingly important in our continued quest
to understand the mechanism of action of complex herbal preparations.
Please note: This article was changed according to the following erratum in Planta
Medica 2012: 78: 302: Reference 31 was changed.
