Synergy Between Chemical Penetration Enhancers

Synergy Between Chemical Penetration Enhancers∗ Keng Wooi Ng†, Wing Man Lau‡, and Adrian C. Williams§ Department of Pharmaceutics, School of Pharmacy, University of Reading, Whiteknights, 226, Reading RG6 6AP, UK Abstract Synergistic combinations of chemical penetration enhancers are important considerations in drug formulation because they can greatly influence transdermal drug delivery, positively or negatively. This chapter describes how synergistic actions between chemical penetration enhancers can be defined, discovered and measured. The chapter considers literature reports where both positive and negative synergism between chemical penetration enhancers have been described. The underlying mechanisms of action that result in enhancer synergy are also discussed. Contents Contents 1 1 Introduction 1.1 Defining Synergy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Discovering Synergy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Quantifying Synergy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 3 4 2 Synergistic Combinations of Penetration Enhancers 2.1 Positive Synergy . . . . . . . . . . . . . . . . . . . . . 2.1.1 Terpenes and Alcohols . . . . . . . . . . . . 2.1.2 Azone and Alcohols . . . . . . . . . . . . . . 2.1.3 Fatty Acids and Alcohols . . . . . . . . . . . 2.1.4 Pyrrolidones . . . . . . . . . . . . . . . . . . . 2.1.5 Esters and Fatty Acids . . . . . . . . . . . . . ∗ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 . 6 . 6 . 6 . 7 . 10 . 10 Self-archived author manuscript. Final version published as: Ng KW, Lau WM, Williams AC (2015) Synergy between chemical penetration enhancers. In: Dragicevic N, Maibach HI (eds) Percutaneous Penetration Enhancers Chemical Methods in Penetration Enhancement: Modification of the Stratum Corneum. Springer Berlin Heidelberg, Berlin, Heidelberg, pp 373–385. DOI: 10.1007/978-3-662-470398_24 † e-mail: [email protected] (current), [email protected] (previous). Current address: School of Pharmacy and Biomolecular Sciences, University of Brighton, Cockcroft Building, Lewes Road, Brighton BN2 4GJ, UK. ‡ e-mail: [email protected] § Corresponding author. e-mail: [email protected] 1 2.1.6 Surfactants . . . . . . . . 2.1.7 Ureas and Alcohols . . . 2.1.8 Surfactants and Alcohols 2.1.9 Sulfoxides and Alcohols . 2.2 Negative Synergy . . . . . . . . . 2.2.1 Amines and Azone . . . . 2.2.2 Amines and Terpenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 11 11 12 12 12 12 3 Conclusions 13 References 13 1 Introduction In transdermal delivery, penetration enhancers are chemicals which improve percutaneous absorption of other molecules. They do so typically by transiently permeabilising the stratum corneum, the outermost layer of the skin which forms the primary barrier to the ingress of exogenous chemicals. Penetration enhancers are particularly useful in facilitating the dermal absorption of active pharmaceutical ingredients. To date, over 300 penetration enhancers have been studied, some of which have been found to act in synergy. Synergy between penetration enhancers presents exciting opportunities for formulation scientists to further improve the bioavailability of transdermal formulations whilst minimising adverse effects and costs. The purpose of this chapter is to provide an overview of synergistic penetration enhancer combinations and to summarise mechanisms of synergy. 1.1 Defining Synergy The term ‘synergy’ is derived from the Greek word, sunergos, which means ‘working together’ (sun-, together; ergon, work). In popular ussage, synergy is defined as the ‘interaction or cooperation of two or more organisations, substances, or other agents to produce a combined effect greater than the sum of their separate effects’ [1]. However, in specialised usage, such as in the scientific literature, a distinction has long been drawn between positive and negative synergy. Positive synergy refers to synergy as it is popularly understood, i.e. magnification of the combinatorial effect. By a pseudo-mathematical metaphor, we may call this relationship 2 + 2 = 5. Conversely, negative synergy refers to attenuation of effect when multiple entities work together (again, pseudo-mathematically, 2 + 2 = 3). Nonetheless, with regard to penetration enhancers, negative synergy has often been ignored, e.g. ‘a true synergistic effect is achieved when the combination of penetration enhancers elicits a greater effect than the individual components used alone’ [2]. As the field has advanced in recent years, the definition has evolved to include negative synergy, complete with mathematical formulae to quantify the effect (see Sect. 1.3). The distinction between positive and negative synergy is important to this discussion because, although positive synergy in penetration enhancement is desirable and rightly deserves our focus, negative synergy may crucially determine coformulation compatibility and thus should not be ignored. Thus, synergy can be defined as ‘the phenomenon whereby multiple penetration enhancers working together results in 2 an enhancement in skin permeability that is greater or less than the sum of enhancements obtained from these penetration enhancers working independently’. Accordingly, we will discuss both positive and negative synergy between penetration enhancers. 1.2 Discovering Synergy The ideal combination of synergistically acting penetration enhancers should be highly potent at enhancing percutaneous absorption but nonirritant. Discovering synergistic combinations of penetration enhancers with these properties presents a significant challenge, since such combinations are relatively rare and also due to technological limitations in their detection. Synergistic combinations of penetration enhancers are difficult to predict. Thus, synergy has often been discovered serendipitously through experimentation. However, it is possible to discover synergy systematically by screening permutations of penetration enhancer mixtures. To do this effectively, a reasonably large library of candidate penetration enhancers is usually necessary, and the speed of the screening technique is of essence given the large number of possible permutations, multiplied by the need to validate these against specific permeants. To illustrate the enormity of the task, it has been estimated that testing random selections of binary penetration enhancer combinations from the ~300 known penetration enhancers, at a modest 25 compositions per pair, will involve the evaluation of approximately a million different formulations [3]. Traditionally, discovery of synergy has used static Franz diffusion cells. However, the process is laborious and time-consuming, as it requires the measurement and comparison of steady-state flux of multiple permeants with various penetration enhancer combinations. Since skin permeation experiments typically take hours or days to complete, with 10–20 experiments running in parallel, the discovery process can be extremely slow. Various degrees of automation can be introduced to improve the speed and manageability of this process, e.g. by using a flow-through diffusion cell [4, 5] or an automated in vitro dermal absorption (AIDA) system [6]. Nonetheless, skin permeation experiments remain the rate-limiting step in these methods. This, coupled with the need for large numbers of skin samples and the time needed to achieve steady-state flux, makes these methods inefficient at screening large libraries of penetration enhancer combinations for synergy [7]. More recently, high-throughput screening methods have been developed to rapidly discover synergistic combinations of penetration enhancers. The in vitro skin impedance-guided highthroughput (INSIGHT) screening method can rapidly assess the potency of 100 penetration enhancer mixtures per square inch of skin sample simultaneously in an array format [8]. It uses electrical conductivity (by measuring its reciprocal, i.e. electrical resistance) across the skin as a surrogate measure of skin permeability. The screening method is estimated to be over 100 times more efficient than the static Franz diffusion cell. In one study, the authors screened a library of over 5000 penetration enhancer mixtures using INSIGHT. Of these mixtures, 2 % were found to be synergistic. The results reiterate the rarity of synergistic penetration enhancer combinations and the value of high-throughput screening technology in their discovery. Only the leading hits from INSIGHT screening were then validated by skin permeation experiments using the static Franz diffusion cell and further tested for safety and bioavailability. This greatly reduces the need for skin permeation experiments that are the bottleneck in the conventional approach. Since 3 electrical conductivity of the skin usually increases with increasing skin permeability, and can be measured instantaneously, this also speeds up the screening process considerably. Moreover, skin permeability estimated this way can be interpreted independently of the mechanism of action of the penetration enhancers, which is often poorly understood. Nevertheless, caution must be exercised to take into account the physicochemical properties of the permeant for which the penetration enhancer formulation is intended [9]. 1.3 Quantifying Synergy Clearly, to quantify synergy between penetration enhancers, penetration enhancement must be assessed. In the scientific literature, penetration enhancement has been reported in a number of ways, giving rise to numerous parameters, variously termed the ‘permeability enhancement’ [10], ‘flux enhancement’ [10], ‘enhancement factor’ [11, 12], ‘enhancement ratio’ [8, 13, 14], ‘enhancement potential’ [15, 16], ‘enhancement index’ [17] and ‘penetration index’ [18]. Despite the varied terminology, calculation of penetration enhancement is straightforward, and the various terms listed above are interchangeable in most cases. Nevertheless, it is important to clearly define the parameter used when reporting penetration enhancement. For a given penetration enhancer (or combination of penetration enhancers), penetration enhancement can be calculated by comparing skin permeability in the presence and absence of the penetration enhancer(s). Skin permeability is usually represented by the permeability coefficient (Kp ). The enhancement ratio (Er ) is the most commonly used parameter in the scientific literature. Er can be described by Eq. 1: Er = Kp, enhancer Kp, control (1) In addition, Ed can be defined as the difference between the permeability coefficients (Eq. 2). This parameter corresponds to the ‘enhancement potential’ defined by Arora et al. [16]. Ed = Kp, enhancer − Kp, control (2) In Eqs. 1 and 2, Kp, enhancer and Kp, control denote the permeability coefficient in the presence and absence of the penetration enhancer(s), respectively. Generally, penetration enhancers increase Kp ; therefore, Er > 1 and Ed > 0. Comparison of flux, J, has also been used instead of Kp where the permeant concentration in the formulation (C) can be considered constant, e.g. when a saturated solution of the permeant is used [10, 15]. This is because, according to Eq. 3 [19], any change in J under such circumstances is attributed solely to a proportional change in Kp : J = Kp C (3) Electrical conductivity across the skin has also been used instead of Kp as a surrogate measure of skin permeability in calculating Er and Ed . Electrical conductivity usually correlates well with Kp determined using static Franz diffusion cells, especially for hydrophilic permeants [20]. Er and Ed obtained using electrical conductivity measurements have shown good correlation (r 2 ∼ 0.9) with each other; thus, both measures carry the same information [16]. However, as Er presents 4 the information as a ‘fold change’ over Kp, control , it is more susceptible than Ed to the intrinsic variability in Kp, control across different skin samples. For brevity, Er and Ed are referred to in the following discussion collectively as the ‘permeability enhancement’, E, without discriminating one from the other. Synergy can be expressed in terms of the permeability enhancement when a combination of penetration enhancers is used, relative to the permeability enhancement when the same penetration enhancers are used independently of each other. For example, consider two penetration enhancers, A and B, which, when used separately at a concentration of Y % (w/v), give permeability enhancements of EAY and EBY , respectively. When A and B are used in combination at a total concentration (A + B) X ,Y of Y % (w/v), the formulation yields a combined permeability enhancement of EA+B . If the weight fraction of A in the mixture is given by X (where X < 1), then synergy, S, can be calculated using Eq. 4 [21]: X ,Y S= EA+B X · EAY + (1 − X ) · EBY (4) Equation 4 represents synergy in a binary system, i.e. one that contains two penetration enhancers. In theory, synergistic interactions may occur between any number of penetration enhancers. For formulations containing more than two enhancers, Eq. 4 can be condensed to Eq. 5 [16]: EY S = Pn i=1 X i · EiY
(5) In Eq. 5, E Y represents the permeability enhancement resulting from the mixture, Y represents the total concentration (in % w/v) of all penetration enhancers in the mixture combined, n is the number of penetration enhancers in the mixture, X i is the weight fraction of the i-th penetration enhancer in the mixture (where each penetration enhancer is an iteration in the iteration series, and i is an integer denoting the iteration index or counter in that series; 1 ≤ i ≤ n) and EiY is the permeability enhancement obtained independently with Y % (w/v) of the i-th penetration enhancer. From Eqs. 4 and 5, positive synergy is indicated if S > 1. Conversely, if S < 1, then negative synergy as indicated. On the other hand, if S = 1, the enhancement effect is additive, i.e. there is no synergy. 2 Synergistic Combinations of Penetration Enhancers Scientific reports of synergy between penetration enhancers are scarce, and among the reports available, research methods vary greatly. Due to the scarcity and heterogeneity of data, generalisation is difficult. For this reason, the following discussion will be presented as a collection of case studies. It should be read with the following caveats in mind: Firstly, whilst it is possible to quantify synergy using Eqs. 4 and 5, most reports of synergy in existing literature have not adopted this quantitative approach. Therefore, synergy will be discussed in the qualitative or quantitative terms as reported by the original authors. Secondly, some penetration enhancers may function as vehicles while simultaneously exerting some penetration-enhancing effects. For example, propylene glycol 5 (PG) is both a cosolvent and a mild penetration enhancer. In this role, it is usually used in large quantities, which may fundamentally alter the composition of the vehicle and thus the thermodynamic activity (e.g. solubility) of the permeant. Most reports of synergy involve such ‘penetration-enhancing vehicles’, but the vehicle effects are not usually decoupled from permeability enhancement. For the purpose of this discussion, ‘penetration-enhancing vehicles’ are considered penetration enhancers in their own right. 2.1 Positive Synergy 2.1.1 Terpenes and Alcohols Terpenes are plant-derived components found in essential oils. As a penetration enhancer, terpenes are commonly formulated in ethanol and/or PG. Such formulations have consistently demonstrated improved permeability enhancements over terpene or alcohol alone. There have been various reports of positive synergy between terpenes and both alcohols in the transdermal delivery of both hydrophilic and lipophilic molecules. In vivo experiments performed in rats have demonstrated synergistic enhancement between 1,8-cineole and ethanol, as well as between l-menthol and ethanol, in the skin penetration of diclofenac sodium from a gel ointment [22]. Similarly, a combination of d-limonene and ethanol promoted the percutaneous absorption of indomethacin from a gel ointment in rats [23]. Yamane et al. [24] reported that formulations of 1,8-cineole, (+)-limonene, nerolidol and menthone, saturated in 80 % PG, enhanced skin permeability to 5fluorouracil (a small hydrophilic molecule) by 24-, 4-, 18- and 21-fold, respectively. The permeability enhancement by terpenes increased with increasing concentrations of PG in the PG-water co-solvent system. Thus, synergistic action between the penetration enhancers was evidenced by the dependence of terpene activity on PG concentration. The mechanism of synergy between terpenes and PG remains unclear. PG does not itself disrupt stratum corneum lipids at normal skin temperature nor does it enhance terpene uptake into the stratum corneum. However, there is evidence to suggest that PG enhances the disruption of stratum corneum lipids by terpenes [25]. On the other hand, differential scanning calorimetry (DSC) and partitioning studies have suggested different mechanisms of action for the terpenes. Evidence produced by Yamane et al. [24] showed that 1-8-cineole, nerolidol and menthone, but not (+)-limonene, disrupted stratum corneum lipids at physiological skin temperature. The enhancement effect of (+)-limonene was attributed to phase separation of the terpene in stratum corneum lipids. These terpene activities, in conjunction with increased drug partitioning caused by PG, were thought to be responsible for the synergistic activity. 2.1.2 Azone and Alcohols Azone, or 1-dodecylazacycloheptan-2-one, is a synthetic penetration enhancer. It permeabilises the stratum corneum probably by inserting itself within the lipid bilayers or through phase separation within the stratum corneum lipids [26]. Goodman and Barry [13] studied the effect of Azone, PG and their combination on the skin penetration of 5-fluorouracil. Azone (2 %, w/w) in PG increased skin 6 permeability nearly 100-fold, compared to 8.6-fold by 3 % (w/v) Azone in an aqueous vehicle (saline containing Tween 20 at 0.1 %, w/v). PG by itself did not significantly alter skin permeability. DSC results suggested that Azone and PG affected different penetration pathways in the stratum corneum—Azone reduced the resistance of the intercellular pathway by disrupting intercellular lipids, whilst PG affected the intracellular pathway. However, how this modulation of the different pathways gave rise to the drastic increase in skin permeability is not clear. Although the exact mechanism of synergy is not yet fully elucidated, others have shown that Azone is most effective when used in combination with PG. Príborský et al. [27] found that, for the transdermal penetration of insulin, a low concentration of Azone (0.1 %) in combination with 40 % PG was most effective. On the other hand replacing PG with polyethylene glycol (PEG), another polar vehicle, inhibited the penetration-enhancing activity of Azone on metronidazole [28]. This was explained by the observation that PEG, unlike PG, did not penetrate the skin readily and thus was unable to increase partitioning of metronidazole into the skin by co-diffusion (as PG probably did). Interestingly, pretreatment of skin samples with Azone followed by administration of trifluorothymidine (the permeant) in a PG vehicle resulted in 50– 100 % greater penetration enhancement compared to treatment with formulations containing both Azone and PG without pretreatment [29]. The observation that fluid uptake in stratum corneum samples immersed in 2 % (w/w) Azone in PG was at least twice that in PG alone [13] also supports the notion that Azone facilitates PG penetration into the skin. 2.1.3 Fatty Acids and Alcohols Fatty acids have been used extensively to enhance the skin penetration of a range of lipophilic and hydrophilic permeants, usually in an alcohol-based vehicle. Fatty acids and alcohols together produce a two-component system comprising a polar solvent and a nonpolar molecule. Such blends of penetration enhancers appear to work particularly well for hydrophilic permeants [2]. Positive synergy between fatty acids and alcohols in such mixtures is well documented. Cooper [30] explored various combinations of alcohols and fatty acids in facilitating the skin penetration of nonpolar permeants. Using salicylic acid as a model permeant, and saturated solutions of the permeant to ensure equal thermodynamic activity, the author observed that a 1:1 molar ratio combination of oleic acid (OA) and PG increased steady-state flux 20-fold, compared to OA and PG alone (which Table 1: The flux, rate constant and enhancement ratio of various tenoxicam formulations across hairless guinea pig skin Formulation Flux (µg·cm−2 ·h−1 ) ka (mg·cm−2 ·min−1/2 ) Eb PG (20 %) OA (15 %) PG-OA (20 %:15 %) 1.74 ± 0.17 10.49 ± 0.88 35.39 ± 2.69 10.44 ± 0.48 8.07 ± 0.28 8.96 ± 0.05 1 6 20 Source: Larrucea et al. [31] Flux and release rate data are mean ± standard error a k denotes the rate constant for tenoxicam release from the gel b E values denote the enhancement ratios and are derived by dividing the flux obtained with the respective formulations by that obtained with PG 7 Table 2: Flux of nicardipine formulated with various penetration enhancers across hairless guinea pig skin Formulation Fluxa (µg·cm−2 ·h−1 ) EPG b EOA b EDMI b PG OA DMI PG-OA (90 %:10 %, v/v) PG-OA-DMI (80 %:10 %:10 %, v/v) 0.58 ± 0.40 2.59c 1.39 ± 1.13 115.15 ± 39.05 471.37 ± 352.24 – – – 199 814 – – – 45 182 – – – – 339 Source: Aboofazeli et al. [15] Flux data are mean ± standard deviation b E values represent flux obtained with solvent blend divided by flux obtained with the neat solvent indicated by the subscript c Standard deviation unavailable a themselves showed similar levels of enhancement). The author also reported synergistic effects between combinations of other alcohols (diethylene glycol, PEG) and fatty acids, but noted that unsaturated fatty acids were more potent penetration enhancers than saturated fatty acids in these combinations. Comparison between the flux of salicyclic acid across human epidermis and a silicone (dimethyl polysiloxane) membrane revealed that the interaction of PG with the skin, rather than solubilisation of the permeant per se, was responsible for the enhancement effect of PG. Johnson et al. [10] investigated the influence of ethanol and linoleic acid on the transdermal permeation of corticosterone across human cadaver skin. Whilst 50 % (v/v) ethanol gave a 40-fold increase in flux over PBS, PBS saturated with linoleic acid did not significantly alter flux. The use of 50 % (v/v) ethanol saturated with linoleic acid resulted in a flux enhancement of 900-fold over the base vehicle, phosphate-buffered saline (PBS). This enhancement was 20-fold that obtained using 50 % (v/v) ethanol in PBS. Thus, synergy between ethanol and linoleic acid was apparent in the 50 % (v/v) ethanol solution saturated with linoleic acid. The authors noted that this formulation increased both skin permeability and the solubility of corticosterone over the base vehicle, whilst in the absence of linoleic acid, 50 % (v/v) ethanol increased corticosterone solubility but reduced skin permeability. Furthermore, Larrucea et al. [31] reported that the flux of tenoxicam (1 %) across rat skin, from a gel containing 15 % OA and 20 % PG, was enhanced markedly compared to gels containing either OA or PG. Data from the study are shown in Table 1. It is noteworthy that flux increased by nearly threefold with the OA-PG formulation, compared with the sum of flux obtained separately with either OA or PG. The release kinetics of tenoxicam from the gel across a cellulose nitrate membrane was determined using static Franz diffusion cells. The release rates were similar for all gels and in fact slightly higher from the gel containing PG but not OA. Therefore, the flux was not limited by tenoxicam release from the gel. Rather, the penetrationenhancing effect of OA was found to be dependent on the concentration of PG. From these observations, it is clear that PG modulated the penetration enhancing of OA; thus, the two components worked together synergistically. Synergy between OA, PG and dimethyl isosorbide (DMI) in a binary or ternary mixture has also been reported [15, 32]. Aboofazeli et al. [15] compared the flux enhancements of saturated nicardipine solutions in PG-OA-DMI (80 %:10 %:10 %, v/v), PG-OA (90 %:10 %, v/v) and each neat solvent alone, through hairless guinea 8 pig skin. Flux enhancements were calculated as the flux obtained with each solvent blend over that obtained with the respective neat solvents. The data are summarised in Table 2. The use of solvent blends resulted in fluxes 45- to >800-fold greater than the use of neat solvents. The ternary mixture (PG-OA-DMI) also gave greater flux enhancements over the binary mixture (PG-OA). Considering PG as the primary vehicle in these mixtures, the addition of OA and DMI had a marginal effect on the solubility of nicardipine compared to neat PG, yet it resulted in disproportionate enhancements in the flux of nicardipine from these mixtures. Similarly, Squillante et al. [32, 33] reported remarkable flux enhancements for nifedipine through hairless mouse skin using a PG-OA-DMI (84 %:10 %:6 %, molar ratio) mixture optimised to achieve a balance between high flux and a short lag time. Funke et al. [34] compared the steady-state flux of the lipophilic molecules, estradiol and an estradiol analogue, through hairless mouse skin, in the presence of various compositions of PG, lauric acid and DMI. In this study, in contrast with other studies [15, 32], DMI was regarded as an inert vehicle. Also, unsaturated solutions of the permeants, at 2 % (w/v), were used. Lauric acid (10 %, v/v in DMI) and PG alone exhibited similar levels of flux enhancement—both improved the steadystate flux by 10-fold for estradiol and by 20-fold for its analogue, compared with DMI. A combination of PG-lauric acid (90 %:10 %, v/v) further enhanced steady-state flux by 22-fold over that achieved by lauric acid (10 %, v/v in DMI) or PG alone. Using this formulation, the steady-state flux through the intact skin was comparable to the steady-state flux through the skin from which the stratum corneum had been removed to allow unhindered permeation. Alternative combinations of PG with DMSO, or DMI with other penetration enhancers (DMSO, dodecanol), did not significantly improve flux. By measuring the steadystate flux of lauric acid and PG while varying their concentrations in the mixture, the authors determined that the steady-state flux of lauric acid and PG was dependent on the concentration of each other and that they mutually enhanced their permeation through the skin. The exact mechanism of synergy between fatty acids and alcohols remains unclear in most cases. However, since alcohols are good solvents for lipophilic molecules and readily penetrate the skin, they may solubilise fatty acids to allow the latter to co-permeate more readily. This is supported, for example, by the observation that the flux of nitroglycerin in aqueous ethanol (< 70 %, v/v) through the skin was directly related to its solubility in ethanol and depended linearly on concomitant ethanol flux [35]. Evidence also suggests that PG may enhance skin permeability by solvating keratin in corneocytes, which may enhance intracellular transport of permeants, in addition to enhanced intercellular transport by other lipid-disrupting penetration enhancers [13]. Additionally, it has been suggested that PG may prolong the retention of OA in the stratum corneum, thus potentiating the effect of OA on stratum corneum lipid disruption [13]. Increased permeation of fatty acids and their lipid-disrupting action on the skin barrier may in turn instigate more rapid alcohol penetration to further propagate the effect. Indeed, Taguchi et al. [36] showed that fatty acids enhanced PG distribution in the skin. Alcohols may also prevent hydrophilic permeants from forming hydrogen bonds within the skin, e.g. by displacing water molecules from binding and nonbinding sites (thus dehydrating it) or by competing with the permeants for hydrogen bonding sites [13]. In the presence of a lipiddisrupting penetration enhancer such as OA, this effect should be significant only if the stratum corneum lipid structure has been disrupted by the penetration enhancer and is no longer limiting the permeant’s rate of diffusion across the skin [37]. 9 2.1.4 Pyrrolidones Pyrrolidones are chemicals containing a lactam ring. Pyrrolidone derivatives such as 2-pyrrolidone, N-methyl-2-pyrrolidone (MP) and 1-lauryl-2-pyrrolidone (LP) have been investigated as penetration enhancers. These are polar, aprotic and colourless solvents at room temperature. Sasaki et al. [38] studied the skin penetration of phenolsulfonphthalein (phenol red) under the influence of MP and LP across rat skin. Phenol red was prepared in isopropyl myristate (IPM) containing up to 2 mmol L−1 of MP and/or LP. In isolation, MP alone had low penetration-enhancing properties, whereas LP had high penetration-enhancing properties but was limited by a prolonged lag time. The steady-state flux obtained with MP-LP (at 2 mmol L−1 of each penetration enhancer) was 1.8-fold the steady-state flux obtained independently with MP and LP combined. The mixture also produced a shorter lag time (0.10 h) compared to MP (0.62 h) or LP (2.14 h) alone. The enhanced steady-state flux was maintained when the concentration of MP was reduced by 75 %, and the short lag time was maintained even with a 95 % reduction in MP concentration. Mechanistic studies revealed that LP enhanced skin penetration of MP, while MP facilitated the accumulation of LP in the skin. 2.1.5 Esters and Fatty Acids The choline ester, lauroyl choline (LC), has been investigated as a penetration enhancer for transdermal, oral and rectal drug delivery [39, 40]. LC (2 %, w/v) has produced positive synergy when used with OA (2 %, v/v) formulated in a neat PG vehicle, in the skin penetration of 17β-estradiol and acyclovir [40]. Whereas the addition of LC and OA resulted in a 6.9-fold and 1.3-fold increase in the permeability of hairless mouse skin to 17β-estradiol over PG alone, addition of both LC and OA to the PG vehicle yielded a 14-fold increase in skin permeability over PG alone. This combined enhancement was nearly twice the sum of independent enhancements attributed to LC-PG and OA-PG. The synergistic effect between LC and OA was even more pronounced in the delivery of acyclovir across hairless mouse skin. Whilst OAPG enhanced skin permeability to acyclovir by 139-fold over PG alone, no detectable amount of acyclovir penetrated the skin when administered with LC-PG. However, LC-OA-PG enhanced skin permeability to acyclovir 404-fold over PG alone or nearly 3 times the sum of the independent effects derived from LC-PG and OA-PG. However, in the same study, a similar co-formulation of LC (5 %, w/v) and OA (5 %, v/v) in PG yielded an additive rather than synergistic effect on the skin penetration of nitroglycerin. It is also not clear from this study whether the apparent synergy between LC and OA can exist independently of PG. 2.1.6 Surfactants Two of the most potent synergistic penetration enhancer combinations identified using the INSIGHT screening method were sodium lauryl ether sulfate (SLES; 0.35 %, w/v) in combination with 1-phenylpiperazine (PPZ; 0.15 %, w/v), and N-lauroyl sarcosine (NLS; 0.6 %, w/v) in combination with sorbitan monolaurate (Span® 20, S20; 0.4 %, w/v). These penetration enhancer mixtures were tested for their ability to enhance the permeation of inulin (a hydrophilic macromolecule) through porcine skin, using static Franz diffusion cells. SLES-PPZ and NLS-S20 yielded permeability 10 enhancements of approximately 80-fold (S = 3.5) and 30-fold (S = 2.5), respectively, while exhibiting low irritation potentials [8]. SLES, NLS and S20 are well-established penetration enhancers in their own right [41, 42, 43]. SLES and NLS (as a sodium salt) are anionic surfactants, whereas S20 is a nonionic surfactant. The lesser known penetration enhancer, PPZ, is a member of the piperazine (nitrogen-containing ring) family of chemicals. Other derivatives of PPZ have been investigated as penetration enhancers, and PPZ itself has been used alone as a penetration enhancer for oral delivery [44]. However, very little information is available about its mechanism of action as a penetration enhancer except that it purportedly enhances permeation through the paracellular route [45]. It is interesting that among the most potent synergistic penetration enhancer mixtures identified using the INSIGHT screening method are three surfactants. Surfactants have multiple documented mechanisms of action. A main function of surfactants is the solubilisation of lipophilic molecules. Thus, in the skin, these surfactants may solubilise and disrupt stratum corneum lipids. Indeed, SLES and the closely related sodium lauryl sulfate (SLS) have both been shown to extract human epidermal lipids in vitro [46]. Surfactants also form micelles above their critical micelle concentrations. SLS and SLES have been shown to permeabilise the skin when used above their critical micelle concentrations [47]; it has been suggested that micelles thus formed may additionally act as carriers to transport permeants across the stratum corneum [48]. In the case of NLSS20, the penetration enhancers formed aggregates suggestive of micelles in the formulation, although whether these aggregates were responsible for permeabilising the skin remains to be investigated [49]. 2.1.7 Ureas and Alcohols Some derivatives of urea (1-dodecylurea, 1,3-didodecylurea and 1,3-diphenylurea) have shown positive synergy with PG when used as saturated solutions (i.e. at approximately 2.6 mg mL−1 , 0.7 mg mL−1 and 15 mg mL−1 , respectively). Williams and Barry [37] applied 5-fluorouracil and the urea-based penetration enhancers, both as saturated solutions, in liquid paraffin or PG. PG alone or the urea analogues (including urea itself) in liquid paraffin had no significant enhancing effect on the skin penetration of 5-fluorouracil. However, when administered in PG, the urea analogues (but not urea itself) enhanced skin permeability to 5-fluorouracil by sixfold compared to either PG alone or the urea analogues in liquid paraffin. Penetration enhancement was attributed mainly to increased diffusivity of the skin (which increased by ninefold), as partitioning of the hydrophilic 5-fluorouracil from the formulation into the skin was reduced by a factor of about 0.7 with the use of PG. PG may however potentiate the effect of the urea-based penetration enhancers, which are lipophilic in nature, by increasing their partitioning into the skin. 2.1.8 Surfactants and Alcohols NLS and ethanol have demonstrated positive synergy in enhancing skin penetration of fluorescein, a hydrophilic and moderately large molecule [50]. In isolation, ethanol and NLS showed little (threefold in the case of ethanol) or negligible effect on skin permeability to fluorescein, the electrical resistance across the skin or the structural organisation within the stratum corneum. However, in combination, the penetration enhancer mixture led to a 24-fold enhancement in fluorescein delivery. 11 The optimal composition of this penetration enhancer mixture was 3 % (w/v) NLS (range studied: 0–3 %, w/v) and 25–50 % (v/v) ethanol (range studied: 0–100 %, v/v). Mechanistic studies revealed that NLS led to significant fluidisation of stratum corneum lipids and induced conformational changes in stratum corneum proteins, whereas ethanol extracted stratum corneum lipids. The authors suggest that stratum corneum lipid extraction by ethanol may increase NLS permeation into the stratum corneum, thus potentiating its effects on stratum corneum lipid fluidity and stratum corneum protein conformation. 2.1.9 Sulfoxides and Alcohols Positive synergy between dimethyl sulfoxide (DMSO) and PG in enhancing percutaneous insulin absorption has been described [27]. Insulin, being a macromolecule, does not normally penetrate the skin to any appreciable degree. Indeed, in the absence of PG and DMSO, insulin penetration across newborn pig skin was extremely low and was undetectable in some instances using the enzyme-linked immunosorbent assay employed. PG (20 %) and 5 % DMSO independently enhanced percutaneous absorption of insulin by approximately threeand sixfold, respectively. In comparison, a mixture of 20 % PG and 5 % DMSO resulted in a nearly 70-fold enhancement over non-enhanced insulin penetration or approximately 7 times the sum of the independent enhancements brought about by 20 % PG and 5 % DMSO. The base vehicle in this case comprised PG and sodium citrate buffer (pH 3.0). The authors did not comment on the mechanism of synergy. 2.2 Negative Synergy 2.2.1 Amines and Azone Wang et al. [14] reported negative synergy between ethanolamine and Azone, as well as between ethanolamine and menthol (Sect. 2.2.2), in a vehicle comprising IPM-ethanol (4:1, w/w). In this study, ethanolamine was used at a 1:1 molar ratio to the permeant, scutellarin, which was used in saturation. Azone was used at 5 % (w/w). The enhancement ratio was calculated from the cumulative amount of scutellarin penetrated across rat skin over 12 h in the presence or absence of the penetration enhancer(s). The results are summarised in Table 3. Ethanolamine and Azone were both effective penetration enhancers when used in isolation. However, in combination, ethanolamine and Azone decreased steady-state flux and increased the lag time, both by approximately twofold, compared to either penetration enhancer alone. This significantly reduced the amount of scutellarin penetrated and thus the enhancement ratio. The permeability coefficient was also reduced by over threefold compared to ethanolamine alone. In this instance, it was proposed that the amines formed ion pairs with the permeant. Such ion pairs altered the physicochemical properties of the permeant, such as lipophilicity, making them more favourable for skin penetration. However, the authors did not discuss potential vehicle effects or possible mechanisms of the synergy. 2.2.2 Amines and Terpenes In the same study described in Sect. 2.2.1, Wang et al. [14] also reported that a combination of ethanolamine and menthol significantly reduced steady-state flux and 12 Table 3: Skin penetration of scutellarin across rat skin Enhancer J a (µg·cm−2 ·h−1 ) Tl b (h −1 ) Kp c (cm·h−1 ) Ed Ethanolamine Azone Menthol Ethanolamine + Azone Ethanolamine + menthol 55.10 ± 6.65 42.40 ± 4.61 27.80 ± 2.04 24.90 ± 2.03 29.80 ± 3.19 1.86 ± 0.85 1.72 ± 0.29 3.51 ± 0.67 3.31 ± 0.17 1.38 ± 0.32 17.50 ± 1.80 0.18 ± 0.02 3.73 ± 0.27 4.81 ± 0.39 16.30 ± 1.75 7.11 5.64 3.28 3.01 4.26 Source: Wang et al. [14] Data are mean ± standard error a J steady-state flux b Tl lag time c Kp permeability coefficient d E enhancement ratio, calculated as the ratio of cumulative amount of scutellarin penetrated over 12 h in the presence of the enhancer to the cumulative amount penetrated in its absence the enhancement ratio compared to ethanolamine alone (Table 3). Meanwhile, the lag time was shortened, whilst the permeability coefficient appeared to be unaffected. Menthol was used at 5 % (w/w) in this study. 3 Conclusions Synergy between penetration enhancers can be harnessed to greatly improve percutaneous absorption without necessitating a large increase in the dose of penetration enhancers used. However, the choice of combination will depend on various formulation considerations, including the mechanism of synergy and the physiochemical properties of the permeant. In designing a formulation containing multiple penetration enhancers, positive as well as negative synergy should be considered to ensure optimal potency, safety and compatibility. However, there are few reports of negative synergy in the existing literature compared to positive synergy. This is probably attributable to some degree to reporting bias arising from the traditional focus on positive synergy. Better documentation of synergy (e.g. more quantitative reporting covering negative as well as positive synergy) will clearly be beneficial. The molecular mechanisms underlying most synergistic combinations of penetration enhancers have also not been fully elucidated. It is interesting to note that, judging from the case studies summarised in this chapter, the majority of synergistic actions reported in the existing literature involve alcohols that are often used as vehicles. This may or may not indicate significant vehicle effects. In any case, a better understanding of the underlying molecular mechanisms may help in the selection of synergistic combinations of penetration enhancers. References [1] Oxford Dictionaries (2010) Synergy. In: Oxford dictionaries. Oxford University Press. http://oxforddictionaries.com/definition/english/synergy. Accessed 13 Sept 2012 13 [2] Møllgaard B (1993) Synergistic effects in percutaneous enhancement. In: Walters KA, Hadgraft J (eds) Pharmaceutical skin penetration enhancement. M. Dekker, New York, pp 229–242 [3] Karande P, Jain A, Mitragotri S (2006) Insights into synergistic interactions in binary mixtures of chemical permeation enhancers for transdermal drug delivery. J Control Release 115:85–93 [4] Bronaugh RL, Stewart RF (1985) Methods for in vitro percutaneous absorption studies IV: the flow-through diffusion cell. J Pharm Sci 74:64–67 [5] Addicks WJ, Flynn GL, Weiner N (1987) Validation of a flow-through diffusion cell for use in transdermal research. Pharm Res 4:337–341 [6] Moody RP, Martineau PA (1990) An automated in vitro dermal absorption procedure: I. Permeation of (14)C-labelled N, N-diethyl-m-toluamide through human skin and effects of short-wave ultraviolet radiation on permeation. Toxicol In Vitro 4:193–199 [7] Karande P, Mitragotri S (2002) High throughput screening of transdermal formulations. Pharm Res 19:655–660 [8] Karande P, Jain A, Mitragotri S (2004) Discovery of transdermal penetration enhancers by high-throughput screening. Nat Biotechnol 22:192–197 [9] Jain A, Karande P, Mitragotri S (2006) High throughput screening of transdermal penetration enhancers: opportunities, methods and applications. In: Smith EW, Maibach HI (eds) Percutaneous penetration enhancers. CRC/Taylor & Francis, Boca Raton, pp 319–334 [10] Johnson ME, Mitragotri S, Patel A, Blankschtein D, Langer R (1996) Synergistic effects of chemical enhancers and therapeutic ultrasound on transdermal drug delivery. J Pharm Sci 85:670–679 [11] Kadir R, Stempler D, Liron Z, Cohen S (1988) Penetration of adesanosine into excised human skin from binary vehicles: the enhancement factor. J Pharm Sci 77:409–413 [12] Saleem M, Bala S, Liyakat, Aeajaz A (2010) Effect of different carriers on in vitro permeation of meloxicam through rat skin. Indian J Pharm Sci 72:710–718 [13] Goodman M, Barry BW (1988) Action of penetration enhancers on human skin as assessed by the permeation of model drugs 5-fluorouracil and estradiol. I. Infinite dose technique. J Invest Dermatol 91:323–327 [14] Wang M, Fang L, Ren C, Li T (2008) Effect of ion-pairing and enhancers on scutellarin skin permeability. J Pharm Pharmacol 60:429–435 [15] Aboofazeli R, Zia H, Needham TE (2002) Transdermal delivery of nicardipine: an approach to in vitro permeation enhancement. Drug Deliv 9:239–247 [16] Arora A, Kisak E, Karande P, Newsam J, Mitragotri S (2010) Multicomponent chemical enhancer formulations for transdermal drug delivery: more is not always better. J Control Release 144:175–180 14 [17] Williams AC, Barry BW (1991) The enhancement index concept applied to terpene penetration enhancers for human skin and model lipophilic (oestradiol) and hydrophilic (5-fluorouracil) drugs. Int J Pharm 74:157–168 [18] Barakat NS (2010) Evaluation of glycofurol-based gel as a new vehicle for topical application of naproxen. AAPS PharmSciTech 11:1138–1146 [19] Williams AC (2003) Transdermal and topical drug delivery from theory to clinical practice. Pharmaceutical Press, London [20] Karande P, Jain A, Mitragotri S (2006) Relationships between skin’s electrical impedance and permeability in the presence of chemical enhancers. J Control Release 110:307–313 [21] Karande P, Mitragotri S (2009) Enhancement of transdermal drug delivery via synergistic action of chemicals. Biochim Biophys Acta 1788:2362–2373 [22] Obata Y, Takayama K, Machida Y, Nagai T (1991) Combined effect of cyclic monoterpenes and ethanol on percutaneous absorption of diclofenac sodium. Drug Des Discov 8:137–144 [23] Okabe H, Takayama K, Ogura A, Nagai T (1989) Effect of limonene and related compounds on the percutaneous absorption of indomethacin. Drug Des Deliv 4:313–321 [24] Yamane MA, Williams AC, Barry BW (1995) Terpene penetration enhancers in propylene glycol/water co-solvent systems: effectiveness and mechanism of action. J Pharm Pharmacol 47:978–989 [25] Cornwell PA, Barry BW, Bouwstra JA, Gooris GS (1996) Modes of action of terpene penetration enhancers in human skin; differential scanning calorimetry, small-angle X-ray diffraction and enhancer uptake studies. Int J Pharm 127:9–26 [26] Williams AC, Barry BW (2004) Penetration enhancers. Adv Drug Deliv Rev 56:603–618 [27] Príborský J, Takayama K, Nagai T, Waitzová D, Elis J (1987) Combination effect of penetration enhancers and propylene glycol on in vitro transdermal absorption of insulin. Drug Des Deliv 2:91–97 [28] Wotton PK, Møllgaard B, Hadgraft J, Hoelgaard A (1985) Vehicle effect on topical drug delivery. III. Effect of Azone on the cutaneous permeation of metronidazole and propylene glycol. Int J Pharm 24:19–26 [29] Sheth NV, Freeman DJ, Higuchi WI, Spruance SL (1986) The influence of Azone, propylene glycol and polyethylene glycol on in vitro skin penetration of trifluorothymidine. Int J Pharm 28:201–209 [30] Cooper ER (1984) Increased skin permeability for lipophilic molecules. J Pharm Sci 73:1153–1156 [31] Larrucea E, Arellano A, Santoyo S, Ygartua P (2001) Combined effect of oleic acid and propylene glycol on the percutaneous penetration of tenoxicam and its retention in the skin. Eur J Pharm Biopharm 52:113–119 15 [32] Squillante E, Maniar A, Needham T, Zia H (1998) Optimization of in vitro nifedipine penetration enhancement through hairless mouse skin. Int J Pharm 169:143–154 [33] Squillante E, Needham T, Maniar A, Kislalioglu S, Zia H (1998) Codiffusion of propylene glycol and dimethyl isosorbide in hairless mouse skin. Eur J Pharm Biopharm 46:265–271 [34] Funke AP, Schiller R, Motzkus HW, Günther C, Müller RH, Lipp R (2002) Transdermal delivery of highly lipophilic drugs: in vitro fluxes of antiestrogens, permeation enhancers, and solvents from liquid formulations. Pharm Res 19:661–668 [35] Berner B, Mazzenga GC, Otte JH, Steffens RJ, Juang RH, Ebert CD (1989) Ethanol: water mutually enhanced transdermal therapeutic system II: skin permeation of ethanol and nitroglycerin. J Pharm Sci 78:402–407 [36] Taguchi K, Fukushima S, Yamaoka Y, Takeuchi Y, Suzuki M (1999) Enhancement of propylene glycol distribution in the skin by high purity cis-unsaturated fatty acids with different alkyl chain lengths having different double bond position. Biol Pharm Bull 22:407–411 [37] Williams AC, Barry BW (1989) Urea analogues in propylene glycol as penetration enhancers in human skin. Int J Pharm 56:43–50 [38] Sasaki H, Kojima M, Nakamura J, Shibasaki J (1990) Enhancing effect of combining two pyrrolidone vehicles on transdermal drug delivery. J Pharm Pharmacol 42:196–199 [39] Alexander J, Fix JA (1989) Enhancement of absorption of drugs from gastrointestinal tract using choline ester salts. US Patent 4,822,773 [40] Loftsson T, Somogyi G, Bodor N (1989) Effect of choline esters and oleic acid on the penetration of acyclovir, estradiol, hydrocortisone, nitroglycerin, retinoic acid and trifluorothymidine across hairless mouse skin in vitro. Acta Pharm Nord 1:279–286 [41] Aioi A, Kuriyama K, Shimizu T, Yoshioka M, Uenoyama S (1993) Effects of vitamin E and squalene on skin irritation of a transdermal absorption enhancer, lauroylsarcosine. Int J Pharm 93:1–6 [42] López A, Llinares F, Cortell C, Herráez M (2000) Comparative enhancer effects of Span20 with Tween20 and Azone on the in vitro percutaneous penetration of compounds with different lipophilicities. Int J Pharm 202:133–140 [43] Trabaris M, Laskin JD, Weisel CP (2012) Effects of temperature, surfactants and skin location on the dermal penetration of haloacetonitriles and chloral hydrate. J Expo Sci Environ Epidemiol 22:393–397 [44] Whitehead K, Karr N, Mitragotri S (2008) Safe and effective permeation enhancers for oral drug delivery. Pharm Res 25:1782–1788 [45] Whitehead KA, University of California (2007) Safe and effective methods for improving the oral delivery of macromolecules. University of California, Santa Barbara 16 [46] Bahl M (1985) ESCA studies on skin lipid removal by solvents and surfactants. J Soc Cosmet Chem 36:287–296 [47] Lu G, Moore DJ (2012) Study of surfactant-skin interactions by skin impedance measurements. Int J Cosmet Sci 34:74–80 [48] Moore PN, Puvvada S, Blankschtein D (2003) Challenging the surfactant monomer skin penetration model: penetration of sodium dodecyl sulfate micelles into the epidermis. J Cosmet Sci 54:29–46 [49] Karande P, Jain A, Arora A, Ho MJ, Mitragotri S (2007) Synergistic effects of chemical enhancers on skin permeability: a case study of sodium lauroylsarcosinate and sorbitan monolaurate. Eur J Pharm Sci 31:1–7 [50] Kim Y-C, Park J-H, Ludovice PJ, Prausnitz MR (2008) Synergistic enhancement of skin permeability by N-lauroylsarcosine and ethanol. Int J Pharm 352:129–138 17