Approximately ten different phytochemicals have been evaluated in various combinations using the Chou-Talalay method Saw et al. Previously, we observed what qualitatively appeared to be a strong synergistic interaction in HaCaT keratinocyte cells, using a model system in which cells were co-treated with electrophilic sulforaphane and a small-molecule diphenol that generates reactive oxygen species through redox cycling, di- tert -butylhydroquinone dtBHQ Bauman et al.
There were two other aspects of the data that strongly suggested a synergistic interaction between the compounds, that is, that sulforaphane and dtBHQ-generated ROS have distinct targets in the pathway.
First, the maximum ARE reporter activation achieved by the combination treatment was more than twice as high as either compound could achieve on its own. Thus, as shown in Figure 2 , dtBHQ-generated reactive oxygen species do not act on Keap1 cysteines, which would have increased Nrf2 protein levels. Rather, they act downstream on as-yet unidentified targets.
In addition to the qualitatively strong synergistic effects observed at most tested concentrations, there were possible indications of antagonism at the lowest tested concentrations of sulforaphane and dtBHQ. In order to quantitatively analyze the data for interactions between sulforaphane and dtBHQ, we examined available methods for their appropriateness for Nrf2 activators in general and for these compounds in particular. In addition to the non-traditional, hormetic nature of the sulforaphane dosing curve, the dtBHQ curve is also atypical, with a small but reproducible suppression of ARE reporter expression, compared to basal levels, at the lowest tested concentration Bauman et al.
First, since the compounds do not have linear dose-response curves with a 0,0 x , y intercept, effect-based methods, including Response Additivity, are not applicable, as shown in Figure 1. First, it requires the dose-effect curve to be fit, or modeled, to a particular equation. In practice, most methods including the Chou-Talalay method fit data to a Hill-slope equation. Other equations can be used to fit the data, but there are limitations to this approach for non-traditional dosing curves, as explained in The Nearest-Neighbor Approach Alternative to Curve-Fitting.
Second, assuming the data can be reasonably modeled, most methods based on Loewe Additivity require a constant potency ratio, shown in Figure 1E for Hill-slope equations. Given these limitations, we developed an R code that applies the time-tested principles of Loewe Additivity but releases the constraint that there must be a constant potency ratio for the two dosing curves.
For non-traditional dosing curves, to circumvent the issues associated with fitting a dosing curve to an equation with a reasonable number of parameters The Nearest-Neighbor Approach Alternative to Curve-Fitting , we introduce a nearest-neighbor approach. Loewe Additivity is based on two primary principles. The first is dose-equivalence —that for a given effect of dose b of drug B, there is an equivalent dose of drug A that gives the same effect. The example in Figure 3A illustrates that for a dose of 2 for drug B, with an effect of 2, there is an equivalent dose of drug A a eq that also has an effect of 2, found by interpolating from the dosing curve for drug A.
In the example, a eq is 0. The predicted additive effect PAE of the combination treatment is thus the effect determined from the total equivalent dose. The dosing curve for drug A is used to determine the predicted additive effect, which in this example is an effect of 4.
If the actual effect of the combination of drug A at a dose of 1 and drug B at a dose of 2 is larger than 4, that is evidence for synergy and an interaction between the two drugs. Any dosing pair with a lower actual effect then the zero interaction-predicted additive effect is antagonistic and also evidence of an interaction between the two drugs.
In other words, the null hypothesis in Loewe Additivity is that there is zero interaction, and the actual effect is the same as the predicted additive effect PAE. If the actual result of the combination treatment is different than the predicted additive effect, the null hypothesis of zero interaction is not supported, and there is evidence of an interaction.
A predicted additive effect PAE is determined by calculating a total equivalent dose using either the dosing curve for drug A A or drug B B , and then interpolating using the same dosing curve to determine the predicted effect.
See text for details. C A predicted additive effect is determined using the dose-effect data for Drug B, here with the nearest-neighbor method. The two nearest points to the effect or dose of interest are used for interpolation. The error in the predicted additive effect is generated by a Monte Carlo simulation in the R code, as described in the Methods. The values given here are for illustrative purposes. The major assumption made by the methods based on Loewe Additivity is that the same predicted additive effect is obtained when interpolating from either dosing curve.
This is the inherent nature of the Loewe Additivity approach. The repercussions of this somewhat uncomfortable dual result are detailed elsewhere Tallarida, ; Geary, In brief, the current widely used versions of Loewe Additivity make the key assumption that the two dosing curves both fit to a Hill-slope type equation with the same Hill coefficient n and that they come to the same maximum value, making them parallel on a log-dose scale Figure 1E Roell et al.
Mathematically, this is described in the following equation, where for all dosing pairs a,b :. In this model, a CI value of 1 results when drug A and drug B have no interaction and the effects are simply additive. As noted by others, this equation is equivalent to the Chou Talalay combination index equation e.
Any point along this line is indicative of additivity, any point below indicates synergy, and any point above indicates antagonism. The widely-used methods based on Loewe Additivity—the Combination Index and the Chou-Talalay method—require for two given drugs a constant or at least reasonably constant potency ratio, and thus a linear isobologram.
However, the potency ratio is often not constant for two drugs under consideration Grabovsky and Tallarida, ; Geary, ; Lederer et al. A non-constant potency ratio results in two predicted additive effects for each dosing combination, as illustrated in Figures 3A,B.
Accordingly, there will be two curved isoboles—two apparently equally valid, but different, predicted results. Loewe himself noted this was a likely outcome Loewe, As posited by others, the fact that the linear isobole assumption fails for many drugs may have gone largely ignored due to the metaphorical descriptions Loewe used Tallarida, , or perhaps due to influential reviews in the field in support of the linear isobologram whose mathematical bases have since been shown to contain errors Geary, Others have developed models for interaction analysis that allow for multiple isoboles Grabovsky and Tallarida, However, these models are also specific for data that fit to a Hill-slope equation, and as such they are not appropriate for data that do not fit well to this equation.
An issue for analyzing interactions of hormetic compounds in particular is the difficulty in fitting them to an equation. To fit them properly, these compounds require dosing curves with many points, and equations with a high number of powers of freedom Zou et al. Obtaining sufficient data points to fully define a hormetic dosing curve can be cumbersome, in particular when working with compound libraries or rare samples. Moreover, the biphasic nature of hormesis can greatly complicate analysis of synergistic interactions.
By definition, a given effect is observed at two distinct doses of a hormetic drug Figure 1F. However, the problem can be simplified if the region of interest in the dosing curve is the initial portion with the lower doses.
In regard to fitting this left-most half of the dosing curve for hormetic data, while a linear or Hill-slope type equation might reasonably fit the given data for any particular Nrf2 activator, it can be time-consuming to test and evaluate various fits to determine whether a given fit is indeed reasonable. In addition, the chosen fit may be arbitrary, i. In general, we wanted a method that was more broadly applicable to any given set of dosing data.
An example is shown in Figure 3C. The scenario is analogous to that in Figure 3B , but instead of a fitted curve, each individual data point is used in the analysis.
This equation is then used to solve for b eq. Similarly, the predicted additive effect is solved from the equation of the line fit to the points that flank the total equivalent dose of B on the x -axis. If needed, the effect of a dose of 1 for drug A can also be determined using the same method even if that dose itself was not tested, choosing points on the A dosing curve that flank 1 to interpolate for the expected effect.
This is because these necessarily cannot be interpolated between two points on the actual dosing curve. In reality, a hormetic dosing curve decreases in effect after reaching its peak. However, dosing combinations that result in an effect higher than any on the individual dosing curve will still be found to be synergistic.
When designing plate layouts for experiments, care was taken to avoid any neighboring well effect of the treatments with dtBHQ, as previously reported for tBHQ Braeuning et al.
Thus, a separate well plate was used for the DMSO vehicle and sulforaphane-only wells, and titrations of dtBHQ were set up from lowest to highest concentration across a plate. Cells were plated in a well plate at 5. The following day, immediately prior to treatment, spent media was replaced with 2 ml of fresh complete media. Treatments were added directly to each corresponding well. Harvested plates were subjected to a freeze-thaw cycle prior to analysis.
Experiments were performed in quadruplicate. Relative units of reporter activation were calculated as the firefly luciferase ARE-driven values divided by the Renilla luciferase values. All data were normalized to vehicle treatment alone. Also found in the Supplementary Material are a document with step-wise instructions, a document explaining the processes the R code follows in analyzing the data, and an Excel file to transform the returned values into a table format.
The surfaces were generated by adding a mesh trace in Chart Studio. In general, the R code follows the process outlined in Figure 3C. To generate error bars for the predicted additive effect that capture the error in the points used for the interpolations and extrapolations for each step, a Monte Carlo method is used, and iterations are conducted for each dosing pair analyzed the default is 5, For each of the iterations, the data is randomly selected using the R function rnorm , based on the average and standard deviation of the y value on the dosing curve.
The 5, returned results are then averaged to give the final predicted additive effect, and the standard deviation is used as the reported error in that value. The returned result based on the means is also returned, and alternatively could be reported as the predicted additive effect.
A quantitative measure of the extent of synergy generated by the R code is the FoldSynergy value, which is the actual effect divided by the predicted additive effect, with associated error propagation.
The Monte Carlo iterations also generate a p -value for whether an effect is synergistic or antagonistic. The p -value for synergy is equal to the number of times that the randomly generated predicted additive effect was less than the actual effect, divided by the number of randomizations.
For example, if a p -value of 0. Brown, C. A genome-wide association analysis of temozolomide response using lymphoblastoid cell lines reveals a clinically relevant association with MGMT. Genomics 22, — Genome-wide association and pharmacological profiling of 29 anticancer agents using lymphoblastoid cell lines.
Pharmacogenomics 15, — Carpenter, D. Understanding the human health effects of chemical mixtures. Cedergreen, N. Quantifying synergy: a systematic review of mixture toxicity studies within environmental toxicology. Chou, T. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Drug combination studies and their synergy quantification using the Chou-Talalay method.
Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Enzyme Regul. Quantitation of the synergistic interaction of edatrexate and cisplatin in vitro. Cancer Chemother. Foucquier, J. Analysis of drug combinations: current methodological landscape.
Fraser, T. An experimental research on the antagonism between the actions of physostigma atropia. The antagonism between the actions of active substances. Geary, N. Understanding synergy. Gessner, P. Morselli, S. Garattini, and S. A straightforward method for the study of drug interactions: an isobolographic analysis primer.
Goldoni, M. A mathematical approach to study combined effects of toxicants in vitro : evaluation of the Bliss independence criterion and the Loewe additivity model. In Vitro 21, — Goutelle, S. The Hill equation: a review of its capabilities in pharmacological modelling.
Grabovsky, Y. Isobolographic analysis for combinations of a full and partial agonist: curved isoboles. Greco, W. The search for synergy: a critical review from a response surface perspective. The search for cytotoxic synergy between anticancer agents: a case of Dorothy and the ruby slippers?
Cancer Inst. Groten, J. Mixtures and interactions. Food Chem. Toxicology of simple and complex mixtures.
Trends Pharmacol. Gutierrez, J. The antimicrobial efficacy of plant essential oil combinations and interactions with food ingredients. Food Microbiol. Antimicrobial activity of plant essential oils using food model media: efficacy, synergistic potential and interactions with food components. Hennessey, V. A Bayesian approach to dose-response assessment and synergy and its application to in vitro dose-response studies. Biometrics 66, — Hertzberg, R. Synergy and other ineffective mixture risk definitions.
Total Environ. Jacobs, D. Nutrients, foods, and dietary patterns as exposures in research: a framework for food synergy. Konkoli, Z. Safe uses of Hill's model: an exact comparison with the Adair-Klotz model. Laetz, C. The synergistic toxicity of pesticide mixtures: implications for risk assessment and the conservation of endangered Pacific salmon. Li, Y. Polychlorinated biphenyls, cytochrome P 1A1 CYP1A1 polymorphisms, and breast cancer risk among African American women and white women in North Carolina: a population-based case-control study.
Breast Cancer Res. Liu, R. Potential synergy of phytochemicals in cancer prevention: mechanism of action. Loewe, S. The problem of synergism and antagonism of combined drugs. Arzneimittelforschung 3, — Antagonisms and antagonists. Effect of combinations: mathematical basis of problem. Google Scholar. Marking, L. Mayer and J. Mumtaz, M. A weight-of-evidence approach for assessing interactions in chemical mixtures. Health 8, — Nielsen, E. Pharmacokinetic-pharmacodynamic modeling of antibacterial drugs.
Olshan, A. Risk of head and neck cancer and the alcohol dehydrogenase 3 genotype. Carcinogenesis 22, 57— Pegram, M. Trastuzumab and chemotherapeutics: drug interactions and synergies. Peters, E. Pharmacogenomics 12, — Prichard, M. A three-dimensional model to analyze drug-drug interactions. Antiviral Res. Scher, S. Opinion on the Toxicity and Assessment of Chemical Mixtures. Schwartz, J. The concentration-response relation between PM 2. Schwartz, S. Hernandez and D.
National Academies Press. Shen, J. Polymorphisms in XRCC1 modify the association between polycyclic aromatic hydrocarbon-DNA adducts, cigarette smoking, dietary antioxidants, and breast cancer risk. Cancer Epidemiol. In a similar study, Yap et al. The fractional inhibitory concentration or sub-inhibitory concentration CBO 0. In order to determine the optimum time required to inhibit the growth of K. Synergism and antagonism between two antimicrobial agents can also be detected via time kill analysis.
Conversely, data obtained via the checkerboard assay suggests that the interactions between CBO and meropenem, are additive, which suggests that time kill analysis is merely an assay that is suitable in the determination of the killing time, but not for differentiating synergistic, additive and antagonistic interaction between compounds. In order to determine the relationship between two antimicrobial agents, an additional method such as the checkerboard assay and the E test method, which are more reliable, should be coupled with time kill analysis [ 7 , 20 , 21 ].
A sudden drop in the number of bacterial colonies treated with sub-inhibitory concentration of meropenem detected at 4th hour in Figure 1 and from 2. The presence of meropenem, even in its sub-inhibitory concentration, exerted significant stress which temporarily inhibited the growth of the bacteria. Furthermore, the weakened bacteria were unable to grow when plated on MH agar plate as solid media provide another challenge for proper growth in their weakened state [ 22 ].
Thus, this explains the absence of colonies at the mentioned time point. Normal growth was resumed at the 5th hour as K. In contrast, no growth was observed in the bacteria culture treated with the combination of CBO and meropenem at their respective sub-inhibitory concentration due to the postulated mode of action of CBO in interfering the bacterial membrane stability and increasing the permeability of outer membrane, which, eventually facilitates the influx of meropenem as previously reported by Yap et al.
Subsequently, zeta-potential measurements, outer membrane permeability assays and scanning electron microscopy studies were performed to further understand and confirm the role of CBO in interfering bacterial membrane stability and permeability.
Furthermore, the results from each assay were compared with those performed in other studies involving synergistic combination of CBO and antibiotics. The zeta potential measurement basically reflects bacterial metabolic state and membrane potentials; the higher the growth rate of bacteria, the more negative the measurement is [ 24 ]. As shown in Figure 3 , cells within the control group had the most negative membrane potential whereas a drastic increment up to 3-fold can be seen in cells treated with CBO and meropenem alone and in combination.
In comparison with the work of Yap et al. The mechanistic action of CBO in membrane potential was reported by Yunbin and colleagues, bacteria treated with CBO displayed a 3- to 5-fold increment in their membrane potential when compared with non-treated bacterial cells [ 25 ].
To verify the integrity of the bacterial membrane, the outer membrane permeability test was carried out using 0. Under physiological circumstances, bacterial cells with functional outer membranes had the ability to prevent low concentrations of SDS from reaching the intracellular region of the cell, preventing cell lysis. However, in the presence of a permeabilizer, which disrupts the membrane permeability, an influx of SDS would occur, accumulating to a certain concentration and thus causes cell lysis [ 23 , 26 ].
The concentration of SDS was optimized and fixed at 0. In addition, the duration of this experiment was also optimized and fixed at 1 h in order to prevent cell damage due to prolonged exposure to SDS. Overall, the presence of CBO at 0. This can be observed in the decreased OD when exposed to 0. According to a study conducted by Yap et al. Another similar study which looked at the synergistic relationship between lavender oil and piperacillin also showed an identical trend whereby the presence of essential oil causes the influx of SDS leading to cell death [ 16 ].
This further validates that synergism and additivity, perhaps shared similar severity in terms of bacterial membrane disruption. Scanning electron microscopy was performed on all four groups of cells to observe the effect on membrane integrity of additive interaction between CBO and meropenem Figure 5.
In the presence of CBO alone, bacterial membranes were corrugated and deformed. However, the addition of meropenem caused a more severe corrugation and deformation to the cell morphology. Comparable observations were obtained from other studies involving synergistic combination of essential oil and antibiotic [ 15 , 16 , 27 ].
However, the scanning electron microscopy was only able to access the membrane disruption ability via qualitative means. Thus, comparison between synergism and additivity was further quantified and validated via zeta potential measurement and outer membrane permeability assay mentioned above.
Additivity interactions do have its advantages; in this case, very low concentrations of CBO are required to achieve additivity with meropenem.
As reported by Yap et al. In addition, according to the additive combination reported in this study, CBO with the highest degree of additivity at 0. This suggests that, even in their additive state, only low concentrations of MO, PO and TTO is required to reduce the effective dosage of meropenem significantly and a similar observation have been observed in other reports mentioned earlier [ 9 , 10 , 11 , 12 , 13 ].
Additionally, evidence of low concentration used in a crude extract, such as essential oil indicates that the dosage of the individual or group compound within the essential oil needed to achieve additivity is even lower than the predecessor. Essential oils such as CBO are generally regarded as safe by the U. Food and Drug Administration, however, high dosages of essential oil would still cause toxicity in humans [ 28 , 29 ].
Thus it is important that the adverse effects of essential oils be investigated before clinical trials. As such, toxicity evaluations would be performed on minimal numbers of compounds instead of the crude essential oil having a larger variety of compounds that might contribute to the overall toxicity. Such evaluations would eventually, pave the way into the application of antibiotic-adjuvant combinatory treatment in the clinical setting. In conclusion, our study showed that additivity and synergistic interaction between CBO and antibiotic were comparable in their ability to cause bacterial membrane disruption, via comparison with similar studies carried out previously.
Quantitative assessment of the membrane disruption ability between CBO and meropenem via zeta potential measurement further confirmed additivity transcended the synergistic combination between CBO and piperacillin. In addition, the outer membrane permeability assay and scanning electron microscopy studies carried out indicated comparable effects of additivity and synergistic interaction between CBO and antibiotics.
Note should be taken our current study focuses particularly on the CBO essential oil, and this should not be assumed to represent the additive interaction as a whole.
However, this preliminary evidence of the significance of additive interaction could be extrapolated to include other essential oil-antibiotic combination as a viable alternative to be supplemented with other treatment strategies in tackling antibiotic resistance. The additive interaction may not be able to substitute synergistic interactions completely; however, this could be a stimulus in propelling the efficacy of combinatory therapy in a new frontier.
Perhaps in the future, additivity may actually transcend synergism. Cinnamon bark Cinnamomum verum , marjoram Origanum majorana , peppermint Mentha x piperita and tea tree Melaleuca alternifolia essential oils used throughout the studies were purchased from Aroma Trading Ltd. Milton Keynes, UK. Meropenem was purchased from Sigma-Aldrich Corporation St. The MICs of essential oils and meropenem were determined qualitatively and quantitatively via the color change in resazurin and relative fluorescent unit.
The checkerboard assay was performed as detailed by Lorian with slight modifications, as described in the REMA [ 7 ]. Ten serial, two-fold dilutions of meropenem and five serial, two-fold dilutions of essential oils were prepared to determine the combinatory effects of essential oils and meropenem against K. Combinatory relationship between CBO and meropenem was expressed in terms of fractional inhibitory concentration index FICI using the following formulas [ 6 , 8 ]:.
The essential oil which yielded the highest value of FICIc in the additive range was used for subsequent assays. The test concentration of essential oil and meropenem used were determined from the checkerboard assay with combination yielding the highest FICI in the additive range, among the four tested essential oil, which is cinnamon bark essential oil CBO.
Immediately after inoculation, viable counting was performed every four hourly until 20 h. In the event of rapid killing, the measurement for viable counting was recorded every half hourly. The time kill analysis was performed in triplicates. The zeta potential of non-treated and treated K. The treatment time for all the treatment groups were as determined in the time kill analysis whereas the concentration of CBO and meropenem used were as determined from the checkerboard assay.
Add a badge to your website or intranet so your workers can quickly find answers to their health and safety questions. Although every effort is made to ensure the accuracy, currency and completeness of the information, CCOHS does not guarantee, warrant, represent or undertake that the information provided is correct, accurate or current.
CCOHS is not liable for any loss, claim, or demand arising directly or indirectly from any use or reliance upon the information. OSH Answers Fact Sheets Easy-to-read, question-and-answer fact sheets covering a wide range of workplace health and safety topics, from hazards to diseases to ergonomics to workplace promotion.
Search all fact sheets: Search. Type a word, a phrase, or ask a question. In addition to synergism, other terms are used to define toxicologic interactions. There are various examples including: a Carbon tetrachloride and ethanol ethyl alcohol are individually toxic to the liver, but together they produce much more liver injury than the sum of their individual effects on the liver.
0コメント