|Home | About | Journals | Submit | Contact Us | Français|
Formulation of a new oil-in-water (o/w) microemulsion composed of castor oil/Tween 80/ethanol/phosphate buffer for enhancing the loading capacity of an anti-inflammatory drug piroxicam has been accomplished. The pseudo-ternary phase diagram has been delineated at constant surfactant/cosurfactant ratio (1:2). The internal structure of so created four-component system was elucidated by means of an analysis of isotropic area magnitudes in the phase diagram. Conductivity (σ), kinematic viscosity (kη), and surface tension (γ) studies with the variation in Φw (weight fraction of aqueous phase) show the occurrence of structural changes from water-in-oil (w/o) microemulsion to oil-in-water (o/w). Along with the solubility and partition studies of piroxicam in microemulsion components, the changes in the microstructure of the microemulsion after incorporation of drug have been evaluated using pH, σ, γ, kη, and density studies. Piroxicam, a poorly water-soluble drug displayed high solubility (1.0%) in an optimum microemulsion formulation using ethanol (55.0%), Tween 80 (26.5%), castor oil (7.5%), and phosphate buffer (11.0%). The results have shown that the microemulsion remained stable after the incorporation of piroxicam. Fluorescence spectra analysis taking pyrene as fluorescent probe was performed, and the results showed that pyrene was completely solubilized in the oil phases of the bicontinuous microemulsions. The fluorescence spectrum of the model drug piroxicam was used to probe the intramicellar region of nonionic microemulsion. The results showed that the piroxicam was localized in the interfacial film of microemulsion systems more deeply in the palisade layer with ethanol as the cosurfactant.
Piroxicam is a nonsteroid anti-inflammatory compound with analgesic and antipyretic effects, used for the treatment of rheumatoid arthritis, osteoarthritis, and traumatic contusions. However, it has been associated with gastrointestinal side effects. It is possible to minimize these problems by developing drug carriers to prevent the direct contact of drug with gastric mucosal or that allow the topical administration of drug (1,2).
Microemulsions are optically isotropic, transparent, and thermodynamically stable homogeneous solutions of oil and water, stabilized by addition of a surfactant and usually a cosurfactant (3,4). These structures have been considerably investigated as drug delivery and carrier system for a wide range of drugs, including analgesics and anti-inflammatory, and also used to dissolve lipophilic drugs in aqueous medium or hydrophilic drugs in lipophilic medium (4,5). Oil-in-water microemulsions have been described as a reservoir system that can inhibit drug release, increasing the topical effect (6). Several mechanisms have been proposed to explain the advantages of microemulsion or the transdermal delivery of drugs (7). First, a large amount of drug can be incorporated in the formulation due to the high solubilizing capacity, with increased thermodynamic activity toward the skin. Second, the permeation rate of a drug from microemulsion may be increased, since the affinity of the drug to the internal phase in microemulsion can be easily modified, to favor partitioning, using different internal phases and changing the composition of the microemulsion. Third, the surfactant and cosurfactant used in the microemulsion may reduce the various diffusional barriers by acting as penetration enhancers (8,9).
For the selection of components of a biocompatible microemulsion system, the use of nonionic surfactants has been widely accepted, since these are compatible and retain its utility over a broad range of pH values and may affect the skin barrier function (10–12).
Microemulsion comprises different structures (water-in-oil (w/o), oil-in-water (o/w), and bicontinuous) and these help in releasing the drug (13,14). It is necessary to characterize the microstructure of pure and drug-loaded microemulsion. The changes in the internal structure of a microemulsion can be monitored by analyzing conductivity, viscosity, density, surface tension, the fluorescence probe studies, etc. (15–17). The incorporated drug may or may not influence the microstructure. o/w and w/o microemulsions may show different behavior for the release of both hydrophilic and lipophilic drugs.
In the present work, an attempt has been made to construct a microemulsion system for poorly water-soluble nonsteroid anti-inflammatory drug piroxicam, comprising of castor oil, a nonionic surfactant Tween 80, a short-chain alkanol cosurfactant (ethanol), and phosphate buffer (PB) pH 7.4. The pseudo-ternary phase diagram has been constructed for the chosen system at a constant surfactant/cosurfactant ratio (1:2). The reason for the selection of the particular oil chosen was that the castor oil has a hydroxyl group in addition to unsaturation, making it more polar. Ricinoleic acid is the main component of castor oil and it exerts anti-inflammatory effects (18). Polyoxyethylene fatty acid, stearic acid, and oleic acid are used in emulsifiers in oil/water type creams and lotions.
Conductivity, viscosity, surface tension, and the fluorescence behavior of the pyrene are employed to investigate the gradual changes occurring in the microstructure of microemulsion. Pyrene is popular fluorescent probe, which is used to study the microheterogeneous media. The fluorescence spectrum of pyrene was used to sense the micropolarity of the o/w microemulsion. In addition, how stability, optical texture, and microstructure of microemulsion formulation are influenced by piroxicam is analyzed. To improve the solubility of piroxicam, an effort has been made to develop an optimum o/w microemulsion. It is therefore expected that the use of microemulsion formulation may enhance the solubility of piroxicam and prevent its degradation.
Tween 80 (polyoxyethylene sorbitan monooleate), absolute ethanol (≥99.8%), and castor oil were purchased from Fluka. Pyrene (98%) was purchased from Sigma-Aldrich. Piroxicam was generously provided by Amson Vaccines & Pharma (PVT) Ltd and used without further purification. Phosphate buffer (0.01 M, pH 7.4) was used as the hydrophilic phase. Buffers were prepared using NaH2PO4/Na2HPO4. 0.1 M NaOH and HCl were used to maintain the pH of the solution.
The pseudo-ternary phase diagram was mapped (as shown in Fig. 1) using oil (castor oil), surfactant (Tween 80; HLB=15), cosurfactant (ethanol), and aqueous phase PB (pH 7.4) at 25±0.01°C with constant surfactant/cosurfactant mass ratio (1:2). The temperature was kept at 25±0.01°C and was maintained by a Lauda M-20 thermostat. Castor oil was first mixed with Tween 80/ethanol mixture; PB was then added to obtain the desired microemulsion compositions. Transparent, single-phase mixtures were designated as microemulsions. All the samples were stable for over 10 months, remaining clear and transparent.
Eight microemulsions differing from each other by Φw were selected from the single-phase region of phase diagram (Fig. 2) with compositions mentioned in Table I to study their potential as drug delivery system. All of them show stability over 10 months and remain clear and transparent. Piroxicam was dissolved into the pre-weight oil component of the system at a concentration of 1% (w/w) under stirring followed by addition of remaining components.
The homogeneity and optical isotropy of pure and drug-loaded microemulsions were examined by a Polarimeter (ATAGO, AP-100 Automatic Polarimeter) and visual examination at room temperature.
Thermodynamic stability of pure and drug-loaded microemulsions was tested by carrying out centrifugation at 5,500 rpm for 20 min using (Hermle Z200) centrifuge.
Surface tension measurements were made at 25±0.01°C under atmospheric pressure by Torsion Balance (White Elec. Inst. Co. Ltd.) equipped with a ring having circumference of 4.0 cm. The experimental error was about ±0.05 mN m−1.
Densities and specific gravity of pure and drug-loaded microemulsions were measured by making use of an Anton Paar (Model DMA 5000) density meter at 25±0.01°C. The density meter was calibrated before and after each set of density measurement using the density of air and pure water.
The refractive indices of the formulations were determined using a refractometer (ATAGO, RX-5000) by placing one drop of solution on the slide.
The apparent pH of all the selected microemulsions and the drug-loaded microemulsion was determined using a pH Meter (WTW 82362 Weilheim) fitted with a pH electrode (WTW A061414035). The temperature was maintained at 25±0.01°C by a Lauda M-20 thermostat.
The effect of the amount of water phase of microemulsion was monitored quantitatively by measuring the electrical conductivity. The electric conductivity (σ) was measured by means of a Microprocessor Conductivity Meter (WTW 82362 Weilheim) fitted with an electrode (WTW 06140418) having a cell constant of 1.0 cm−1. The temperature was kept at 25±0.01°C and was maintained by a Lauda M-20 thermostat. Conductivity measurements were carried out by titration of oil, surfactant/cosurfactant mixture with buffer (along the dilution line AB in Fig. 1). Further the conductivity of selected and drug-loaded microemulsions was also measured. The error limit of conductance measurements was ±0.02 μS cm−1.
Viscosities were measured with calibrated Ubbelhode viscometer at 25±0.1°C. For each measurement, the viscometer was washed, rinsed, and vacuum-dried. To follow the viscous behavior of the microemulsions, flow time was measured for all the selected and drug-loaded microemulsions (1 wt.% drug). The error limit of viscosities measurements was ±3%.
The absorption and steady-state fluorescence spectra were recorded using a Perkin Elmer Lambda 20 spectrophotometer and a Perkin Elmer LS 55 luminescence spectrometer, respectively, both with an external temperature controlled cell holder at a temperature of 25.0±0.1°C. The fluorescence emission spectrum of pyrene (excitation at 340 nm) was used to obtain the ratio of intensities of the first to the third vibronic peaks (I1/I3). Good resolution of the bands was obtained at the slit width (ex. 5.0 nm, em. 5.0 nm). The scan range used was from 350 to 500 nm. The Photo Multiplier tube voltage was kept at 665 V. The concentration of pyrene was 1.0 μM. The intensities for I1 and I3 are taken at 373 and 384 nm, respectively. The fluorescence emission spectrum of piroxicam at λexc 370 nm was obtained where the emission and excitation slits were fixed at 7.0 nm. The scan range used was from 390 to 650 nm. The concentration of piroxicam was 10.0 μM.
To quantify the solubilization of piroxicam in micellar media of Tween 80–ethanol system, differential absorbance measurements were made in such a way that drug (piroxicam) solution of a particular concentration (1.0×10−5 M) was kept on reference side and the Tween 80–ethanol–piroxicam solution on the sample side in the spectrophotometer.
Oil/buffer partition coefficient was determined by dissolving 20 mg piroxicam in 2 mL castor oil. Buffer was added in 1:1 ratio (v/v). The mixture was shaken for 10 min and centrifuged for 2 h. The two layers were separated and the content of piroxicam in aqueous layer (PB) was assayed by UV–Visible spectrophotometer at 371 nm. The final content of drug in the lipophilic phase was calculated by subtracting the content of piroxicam in aqueous phase from initial loaded content of drug in the lipophilic phase. Further, the effect of the presence of Tween 80 and ethanol on the partition of piroxicam in oil/buffer was studied by adding 5% (w/v) of each Tween 80 and ethanol.
In the present system, microemulsion was prepared using castor oil (fatty acid), which induces highly permeable pathways in the stratum corneum (18–20). Tween 80 is a widely accepted nonionic surfactant, used in many pharmaceutical formulations (21–23). The cosurfactant (ethanol) is used to study the one phase microemulsion region. The presence of alcohol overcomes the need for any additional input of energy. These properties make the components useful as vehicles for drug delivery (24–26).
In the absence of aqueous phase, a solution-like oily phase consisting only of surfactant, oil, and ethanol exists. Ethanol interacts with the ethoxylated head groups of the Tween 80 by hydrogen bonding and affects its critical packing parameter. When water is progressively added to the concentrate it facilitates the organization of the hydrated head groups of the surfactant into a polar core, while the fatty acid tails are immersed in the oil continuous phase. The ethanol suppresses formation of lyotropic liquid crystals. Any free aqueous phase is entrapped in the microstructures. Thus, w/o microstructures are formed. Upon further dilution, the reversed nanostructures grow and convert into a bicontinuous phase and finally invert into o/w microstructures without phase separation.
Figure 1 shows the pseudo-ternary phase diagram and area of existence of microemulsion for Tween 80/ethanol/castor oil/phosphate buffer. Microemulsion in the present study formed spontaneously at ambient temperature when their components were brought in contact.
Phase behavior investigations of this system demonstrated the suitable approach of determining the water phase, oil phase, surfactant concentration, and cosurfactant concentration with which the transparent, one-phase, low-viscous microemulsion system was formed. The phase behavior, as shown by Fig. 1, manifests a two-phase region, a three-phase region, and a large single-phase region, which gradually and continuously transformed from buffer-rich side of binary solution (buffer/surfactant micellar phase) of pseudo-ternary phase diagram toward the oil-rich region. This stresses a continuous transition from water-rich compositions to oil swollen micelles.
The phase study revealed that the maximum proportion of oil was incorporated in microemulsion systems when the surfactant-to-cosurfactant ratio was 1:2. From a formulation viewpoint, the increased oil content in microemulsions may provide a greater opportunity for the solubilization of piroxicam. Eight microemulsions (1–8) were selected from the single-phase isotropic region (Fig. 2), with compositions mentioned in Table I. Selected microemulsion (ME) was further analyzed by conductivity, viscosity, density, surface tension, refractive index, and pH. The values of measured parameters have been presented in Table II.
Conductometry is a useful tool to assess microemulsion structure. Conductivity studies have explained the existence of a characteristic zone with an isotropic microemulsion domain in a continuum. For the determination of electric conductivity (σ) as a function of weight fraction of aqueous component Φw (wt.%) for the oil, surfactant/cosurfactant mixture along the dilution line AB (shown in Fig. 2) has been carried out. The results of variation of σ versus Φw (wt.%) are shown in Fig. 3a. The behavior exhibits profile characteristic of percolative conductivity (27–29). The conductivity is initially low in an oil–surfactant mixture but increases with increase in aqueous phase.
As the volume fraction of water increases, the electrical conductivity of the system slightly increases as well, until the critical Φw is reached. At this stage, a sudden increase in conductivity is observed. This phenomenon is known as percolation, and the critical Φw at which it occurs is known as percolation threshold Φp (27).
The value of conductivity below Φp suggests that the reverse droplets are discrete (forming w/o microemulsion) and have little interaction. Above Φp the value of σ increases linearly and steeply till it touches the value of Kb. The interaction between the aqueous domains becomes progressively more important and forms a network of conductive channel (bicontinuous microemulsion) (30).
Rapid increase in conductivity beyond the percolation threshold (Φp≈6%) up to approximate value of 20% of Φw indicates the existence of network of conductive channels, which corresponds to the formation of water cylinders or channels in an oil phase due to the attractive interactions between the spherical microdroplets of water phase in the w/o microemulsion.
Increasing water content above Φb (Φw>20%), the σ shows a dip in the measured values, which may be due to strong attractive forces as system becomes more viscous (16,30). Figure 3b depicts the variation of log σ versus weight fraction of water (Φw). The change in the slope of log σ can be attributed to the structural transition to bicontinuous from w/o (23), nearly at Φw=6%. The transition takes place once the aqueous phase becomes continuous phase, i.e., at Φb. This is in line with the observation made in phase study. Figure 3a illustrates occurrence of three different structures (namely w/o, bicontinuous, o/w). The conductivity of the microemulsions containing more than 20 wt.% water decreased significantly, probably due to the higher viscosity.
The percolation threshold can be determined from the plot (dσ/dΦw), as a function of the water weight fraction, Φw (wt.%) (30). A maximum in the first derivative of conductance Φw at ~12 wt.% water is observed (Fig. 4), confirming the presence of percolation behavior (bicontinuous microstructure) in this region (31). The electric conductivity of pure selected and drug-loaded microemulsion (1.0%) is given in Table II. A comparison of two systems shows that drug incorporation does not affect the microstructure of the microemulsion.
To avoid the ambiguity of non-Newtonian flow behavior of microemulsion, the flow time has been used as an index of viscosity (32). Flow time of oil, surfactant/cosurfactant mixture along the dilution line AB, as shown in Fig. 2, was measured as a function of weight fraction of water Φw (wt.%) and is shown in Fig. 5. There is a consistent increase in flow time with the increase in the Φw as depicted by Fig. 5. Increase in flow time can be correlated to the increasing size of the droplet.
Similar trend has been observed for the viscosity of oil, surfactant/cosurfactant mixture as a function of Φw (Fig. 6). The rapid change in the viscosity is probably due to the change in the microstructure of the microemulsion. The change in the internal structure could be due to either the change in the shape of droplets or may be due to the transition from w/o to bicontinuous microemulsion. It is well known that increase of volume fraction of dispersed phase in microemulsion increases viscosity of the system (33). For the system studied viscosity increases with increase in Φw (wt.% of aqueous phase).
Difference in the viscosities is more profound for lower water content values in comparison to the dilute system. The microemulsion system is turning to be more viscous with addition of water and thus may help in the slow diffusing of drug at infinite dilution. The microemulsion system thus shows a structural change from oil continuous system to water continuous, which has higher viscosities than the former (34). The plots of kη (kinematic viscosity), d2η/d2Φw and 1/η dη/dΦwversus Φw reflect that the transition occurs at ~11% weight fraction of aqueous phase as shown by Fig. 6. The transition point of surface tension, conductivity, and viscosity plots coincides well at ~11% weight fraction of aqueous phase and confirms the presence of percolative behavior.
Figure 7 demonstrates that the surface tension increases linearly over the same range of water content, but two breaks (at ~7.0 and ~20 wt.% water) suggest that structure changes occur at these compositions. The surface tension measurements showed increment, when measured as a function of weight fraction of aqueous component, except for the ~12% weight fraction where the value suddenly decreased and thereafter a regular increase was observed. This low surface tension value showed the presence of bicontinuous microemulsion between oil- and water-rich system, which is because of the presence of self-assembled organize microstructure in it (14,35). The results coincide well with the electric conductivity and viscosity measurements. It can be assumed that the added alcohol (ethanol) is incorporated in the interfacial structure in such a way that more water is on the outside of the “oil drops”, causing the increase in surface tension. Incorporation of drug showed a negligible change in the surface tension measurements, therefore indicting the possibility of piroxicam molecules into the palisade layer on the inner side of microemulsion.
In the case of oil-in-water microemulsions, the steady-state fluorescence technique was successfully applied. Fluorescence measurements of the hydrophobic probe mainly depend on the polarity of the medium, and hence in bicontinuous microemulsions, it is a good indication of the polarity of the microenvironment in the microemulsion structure (36,37). The fluorescence spectra for pyrene molecule in water, individual oil phase, in alcohols (i.e., ethanol and methanol), and in ethanol/oil are shown in Fig. 8a.
There are four principal vibronic bands in the fluorescence spectrum shown in Fig. 8a and are labeled I to IV. The peak intensity ratio I1/I3 in the steady-state fluorescence spectra is a measurement of the relative polarity of pyrene’s environment (17). Since pyrene reactant is substantially more soluble in oil phases, I1/I3 is expected to be lower in these phases (38). In the present work, for oil phase, the I1/I3 value is 0.68. In relatively polar methanol and ethanol media, I1/I3 values were found to be 1.20 and 1.09, respectively. Water is a highly polar solvent; the solubility of pyrene in this solvent is less than 2 μM. Hence the possibility of formation of excimer leading to I3 signal is extremely low in water. Experimentally I1/I3 value is 1.70 was obtained for this medium.
Figure 8b shows the fluorescence spectra of all the compositions of microemulsion when pyrene is incorporated in their solutions. An overall increase in the intensity of all the four peaks was observed for 2% to 16% of water fraction. The reason might be that as we increase the water fraction, more and more pyrene is partitioned into the oil phase, and hence an increase in the fluorescence intensity is observed.
Plot of I1/I3versus weight fraction of aqueous component composition in microemulsion is shown in Fig. 9. The value of I1/I3 varies between 0.85 and 0.91, which is comparable to a change from oil to water (0.68 and 1.70, respectively).
The I1/I3 fluorescence ratios of pyrene strongly suggest that this probe resides in microenvironments of polarity much lower (oil phase) than that of water or alcohol (39). The polarities of these microphases are similar to those of cosurfactant/oil mixtures (0.94). The following generalizations may be made regarding the fluorescence probe behavior in bicontinuous microemulsions. The I1/I3 values obtained by fluorescence measurements for all the stable bicontinuous microemulsions are closer to 0.88. These results suggest that pyrene is efficiently segregated from the water phase (40). The I1/I3 values in bicontinuous microemulsions systems are closer to the respective pure oil phase. This is due to complete solubility of pyrene in oil phases of the bicontinuous microemulsions. We conclude that all the microemulsions have separate oil microphases, in which pyrene resides.
The fluorescence spectra for piroxicam molecule in water, individual oil phases, in surfactant/cosurfactant mixture (1:2), and in the optimum microemulsion system are shown in Fig. 10.
For oil phase the emission maxima (λem) is 465 nm. In S/CoS (1:2) system λem is 451 nm. Water is a highly polar solvent; the solubility of piroxicam in this solvent is lower than 10 μM. The λem of piroxicam in water is 442 nm. The emission maximum in bicontinuous microemulsion system is (462 nm) closer to the respective pure oil phase. The results showed that the piroxicam was localized in the interfacial film of microemulsion systems more deeply in the palisade layer.
Partition coefficients influence drug transport characteristics, which involve drug absorption, retention, distribution, and elimination. Since drugs are distributed by the blood, they must penetrate and traverse many cells to reach the site of action. Hence, partition coefficients will determine what tissues a given compound can reach.
The partition coefficient (log p) of piroxicam in oil/buffer is 5.03±0.20. The presence of ethanol (5% in buffer) does not affect the partition coefficient (data shown in Table III) whereas Tween 80 (5% in buffer) reduces the log p. The presence of surfactant reduces the concentration of drug in oil. Thus, solubility and partition studies indicate that piroxicam may be present at interface. The drug is entering into the palisade layer on the inner side of droplet, which may help to increase the solubility of piroxicam. The partition coefficients were calculated using Eq. 1 (41);
where A(org) is the absorbance of the organic layer, A(aq) is the absorbance of the aqueous layer, Vf(org) is the final volume of the sample from the organic layer, V(org) is the volume of the aliquot from the organic layer, Vf(aq) is the final volume of the sample from the aqueous layer, and V(aq) is the volume of the aliquot of the aqueous layer.
Figure 11 shows the differential absorption spectra of drug (piroxicam) in the presence of various concentrations of Tween 80 having constant S/CoS ratio (1:2).
The buffer–micelle partition coefficient Kc (dm3 mol−1), a useful parameter to quantify the solubilization of piroxicam in micellar media of Tween 80–ethanol system, can be calculated by using Eq. 2 (42).
Here Ca is the drug concentration (1.0×10−5 M), Csmo represents Cs−CMCо (CMCо is the CMC of Tween 80 in water, i.e., 11.0 μM), and A∞ is the differential absorbance at the infinity of Cs. Kc can be obtained through intercept and slope values of the straight line plot of 1/A against 1/(Ca+Csmo), as shown in Fig. 12. The value of Kc is given in Table IV.
The dimensionless partition coefficient p is related to Kc as p=Kc.nw, where nw is the number of moles of water per dm3 (55.5 mol dm−3), and is reported in Table IV. The standard free energy change of the transfer of additive from bulk water to micelle can be calculated using the following relation (Eq. 3):
Here T is absolute temperature and R is the gas constant. The value of for the piroxicam, using p, is reported in Table IV.
High negative value of indicates the ease of penetration of drug inside the micelles. This is clearly exhibited by the higher values of p and more negative for piroxicam, as shown in Table IV. Tween 80 is nonionic surfactant, and there is no electrostatic interaction, and the hydrogen bonding between the polyoxyethylene groups of Tween 80 and piroxicam makes the complex (Tween 80–piroxicam) more hydrophobic, which corresponds to high value.
The conductivity and viscosity studies along the dilution line (in phase diagram) depict the structural transition from w/o to o/w via bicontinuous phase at ~11% Φw (wt.% fraction of aqueous phase). Among the eight selected microemulsions, ME was found to be optimum for the incorporation of piroxicam. After the incorporation of the drug, microemulsion remained stable and optically clears, with no phase separation. The drug is entering into the palisade layer on the inner side of the droplet, which may result in controlled release of drug. Thus, we can conclude that this microemulsion system helps in increasing the solubility of a highly hydrophobic drug, with the help of hydrophobic component of microemulsion and lipophilic part of surfactant. In addition, the formulation can be explored with high concentration of drug. Results of conductivity, viscosity, density, and surface tension measurements confirm the prediction of a percolation transition to a bicontinuous structure.
The financial support of Quaid-i-Azam University and Higher Education Commission of Pakistan is duly acknowledged.
Asad Muhammad Khan, Phone: +92-519-0642047, Email: email@example.com.
Syed Sakhawat Shah, Phone: +92-519-0642110, Fax: +92-519-0642241, Email: firstname.lastname@example.org.