Bulk and Dispersed Aqueous Phase Behavior of Phytantriol: Effect of ...

9512

Langmuir 2006, 22, 9512-9518

Bulk and Dispersed Aqueous Phase Behavior of Phytantriol: Effect of Vitamin E Acetate and F127 Polymer on Liquid Crystal Nanostructure Yao-Da Dong,† Ian Larson,† Tracey Hanley,‡ and Ben J. Boyd*,† Department of Pharmaceutics, Victoria College of Pharmacy, Monash UniVersity, ParkVille, Victoria 3052, Australia, and Bragg Institute, Australian Nuclear Science and Technology Organisation, Menai, New South Wales 2234, Australia ReceiVed June 14, 2006. In Final Form: August 14, 2006 Phytantriol (3,7,11,15-tetramethylhexadecane-1,2,3-triol, PHYT) is a cosmetic ingredient that exhibits similar lyotropic phase behavior to monoolein (GMO), forming bicontinuous cubic liquid crystalline structures (QII) at low temperatures and reversed hexagonal phase (HII) at higher temperatures in excess water. Despite these similarities, phytantriol has received little attention in the scientific community. In this study, the thermal phase behavior of the binary PHYTwater and ternary PHYT-vitamin E acetate (VitEA)-water systems have been studied and compared with the behavior of the dispersed cubosomes and hexosomes formed with the aid of a stabilizer (Pluronic F127). The phase behavior and nanostructure were studied using crossed polarized light microscopy (CPLM), differential scanning calorimetry (DSC), and small-angle X-ray scattering (SAXS) techniques. The presence of lipophilic VitEA in the PHYT-water system suppressed the temperature of the QII-to-HII-to-L2 transitions, indicating that lipophilic compounds, in relatively small amounts, may have a significant impact on the phase behavior. Increasing the F127 concentration in the phytantriolbased cubosome system did not induce the QII(Pn3m) to QII(Im3m) transition known for the GMO-water system. This indicates a different mode of interaction between F127 and the lipid domains of phytantriol-water systems. Taken together, these results indicate that phytantriol may not only provide an alternative lipid for preparation of liquid crystalline systems in excess water but may also provide access to properties not available when using GMO.

Introduction Nanostructured, lipid-based self-assembled systems, such as lyotropic liquid crystals, are generating substantial interest in the fields of drug,1,2 protein,3 and vaccine delivery. This interest is mainly due to the potential of these systems to incorporate compounds of varying physicochemical properties and to provide a mechanism of sustained release. Hydrophilic drugs can be incorporated into the aqueous domains, and lipophilic drugs can be dissolved in the lipidic regions of the structure. Shah et al.4 and Drummond and Fong5 have provided comprehensive reviews of nondispersed liquid crystals as drug delivery systems. In certain instances, the bulk liquid crystalline structure may be dispersed to form sub-micrometer particles which retain the internal structure of the nondispersed liquid crystal bulk phase.6 In the case of lamellar, hexagonal, and cubic phases, these particles have been termed liposomes, hexosomes, and cubosomes, respectively. However, the ability of the dispersed particle form to provide a sustained release delivery system is yet to be conclusively demonstrated.7 The majority of reported studies on bulk and dispersed liquid crystalline systems, have centered around the glyceride-based lipids, principally monoolein (GMO, Figure 1)8-14 and mono* To whom correspondence should be addressed. E-mail: ben.boyd@ vcp.monash.edu.au. † Department of Pharmaceutics, Victoria College of Pharmacy, Monash University. ‡ Bragg Institute, Australian Nuclear Science and Technology Organisation. (1) Chang, C. M.; Bodmeier, R. J. Pharm. Sci. 1997, 86, 747-752. (2) Chang, C. M.; Bodmeier, R. J. Controlled Release 1997, 46, 215-222. (3) Sadhale, Y.; Shah, J. C. Int. J. Pharm. 1999, 191, 65-74. (4) Shah, J. C.; Sadhale, Y.; Chilukuri, D. M. AdV. Drug DeliVery ReV. 2001, 47, 229-250. (5) Drummond, C. J.; Fong, C. Curr. Opin. Colloid Interface Sci. 1999, 4, 449-456. (6) Larsson, K.; Tiberg, F. Curr. Opin. Colloid Interface Sci. 2005, 9, 365369. (7) Boyd, B. J. Int. J. Pharm. 2003, 260, 239-247.

Figure 1. Chemical structures of glyceryl monooleate (GMO), phytantriol (PHYT), and vitamin E acetate (VitEA).

linolein (MLO)15,16 which both form a bicontinuous cubic phase in excess water. However, the ester-based structure of these lipids may limit their practical application, as ester hydrolysis may lead to chemical instability and disruption of the liquid crystalline structure. In addition, commercially produced sources of these lipids, such as the Myverol 18-99K used in this study, are often (8) Landh, T. J. Phys. Chem. 1994, 98, 8453-8467. (9) Gustafsson, J.; Ljusberg-Wahren, H.; Almgren, M.; Larsson, K. Langmuir 1996, 12, 4611-4613. (10) Gustafsson, J.; Ljusberg-Wahren, H.; Almgren, M.; Larsson, K. Langmuir 1997, 13, 6964-6971. (11) Nakano, M.; Sugita, A.; Matsuoka, H.; Handa, T. Langmuir 2001, 17, 3917-3922. (12) Nakano, M.; Teshigawara, T.; Sugita, A.; Leesajakul, W.; Taniguchi, A.; Kamo, T.; Matsuoka, H.; Handa, T. Langmuir 2002, 18, 9283-9288. (13) Borne, J.; Nylander, T.; Khan, A. Langmuir 2000, 16, 10044-10054. (14) Borne, J.; Nylander, T.; Khan, A. Langmuir 2001, 17, 7742-7751. (15) deCampo, L.; Yaghmur, A.; Sagalowicz, L.; Leser, M. E.; Watzke, H.; Glatter, O. Langmuir 2004, 20, 5254-5261. (16) Yaghmur, A.; de Campo, L.; Sagalowicz, L.; Leser, M. E.; Glatter, O. Langmuir 2005, 21, 569-577.

10.1021/la061706v CCC: $33.50 © 2006 American Chemical Society Published on Web 10/17/2006

Aqueous Phase BehaVior of Phytantriol

complex mixtures of amphiphilic substances varying in chain length and level of purity, making their behavior somewhat difficult to characterize and control. One possible alternative to the glyceride lipids for the preparation of liquid crystalline systems is phytantriol (3,7,11,15tetramethylhexadecane-1,2,3-triol). This lipid comprises a highly branched phytanyl tail with a tri-hydroxy headgroup, and importantly, its structure does not include an ester functional group (Figure 1). Phytantriol is available from a number of commercial sources at 95% nominal purity and thus offers a relatively well-defined lipid to prepare viscous liquid crystal systems. Phytantriol has been reported to exhibit lyotropic phase behavior very similar to that of GMO,17 in particular, it forms a bicontinuous cubic phase in excess water at room temperature and reverse hexagonal phase in excess water at higher temperatures. Phytantriol has also been reported to form cubic phase particles initially in the patent literature,18 and more recently by Barauskas et al., who utilized vitamin E TPGS as a stabilizer and showed that the particles adopt the Im3m mesophase structure by cryo-transmission electron microscopy.19 However the mechanisms of formation, structural characteristics, and transition behavior of phytantriol dispersions have not been investigated. Of particular interest is the use of phytantriol as an alternative amphiphile for the preparation of cubosomes and hexosomes. The relative stability and purity of phytantriol compared to GMO suggests it to be a better candidate for application in drug delivery systems. Investigation of liquid crystalline phases formed by phytantriol in addition to the cubic phase has significant merit, as different phases may exhibit different surface properties, drug loading, or release behavior. The understanding of whether the structure of nondispersed phytantriol bulk phase is retained upon dispersion as cubosomes is limited. The nature of the interaction between the stabilizer Pluronic F127 and phytantriol in terms of colloidal stability and phase structure is also unknown. Therefore, in this investigation, the first aim was to determine whether the bulk liquid crystalline nanostructure is retained upon formation of dispersed particles of phytantriol. Second, a study of the properties and stability of particles of dispersed phase and comparison of that with GMO cubosomes was also undertaken. Finally, the means to manipulate the phase behavior and nanostructure of the bulk liquid crystalline phase and the dispersed particles based on phytantriol were investigated. Vitamin E acetate (VitEA) was utilized as an additive in order to manipulate the QII-to-HII phase transition, based on previous observations in this lab that it suppressed the temperature of this transition for the phytantriol/water system.20 Alteration of the phase structure and identity with changes in temperature and composition were determined using small-angle X-ray scattering (SAXS), crossed polarized light microscopy (CPLM), and differential scanning calorimetry (DSC). Materials and Methods Materials. Phytantriol was purchased from Roche (GrenzachWyhlen, Germany) with nominal purity of >96.6% purity (GC assay from certificate of analysis no. 01444062). Vitamin E acetate was purchased from Sigma Aldrich Chemie (Steinhiem, Germany). Myverol 18-99K was donated by Kerry Bio-Science (Norwich, NY). The major components in Myverol 18-99K are GMO (60.9%) and MLO (21.0%).21 Pluronic F-127 was purchased from BASF (Florham Park, NJ). These chemicals were used as received without further (17) Barauskas, J.; Landh, T. Langmuir 2003, 19, 9562-9565. (18) Ribier, A.; Biatry, B. U.S. Patent 5,834,013; November 10, 1998. (19) Barauskas, J.; Johnsson, M.; Tiberg, F. Nano Lett. 2005, 5, 1615-1619. (20) Unpublished results from a manuscript in preparation by Nguyen, T.-H.; Larson, I.; Porter, C. J. H.; Boyd, B. J.

Langmuir, Vol. 22, No. 23, 2006 9513 purification. MilliQ grade (0.05 µS cm-1 at 25 °C) water purified through a Millipore system (Sydney, Australia) was used throughout this study. Glass capillaries for SAXS experiments were purchased from Charles Supper (Natick, MA). Preparation of Nondispersed Liquid Crystalline Systems (“Bulk Phase”). Samples of bulk phase were prepared by weighing appropriate amounts of lipid and water into glass HPLC vials, the samples were mixed by repeated cycles of heating, vortex mixing, and centrifugation. The samples were then heated to form the lowviscosity L2 phase, vortex mixed once more, and injected into glass capillaries. The loaded capillaries were then heat-sealed and left at room temperature for at least 1 week to allow for equilibration. For bulk phases containing both phytantriol and vitamin E acetate, the two lipids were mixed via a roller mixer for at least 1 week prior to the addition of water. Preparation of Dispersed Systems and Particle Size. Unless stated otherwise, 900 mg of lipid was weighed into a 20 mL glass vial. Water (10 mL) containing F127 (1% w/w) was added immediately prior to dispersion by ultrasonication (Misonix XL2000, Misonix Incorporated, Farmingdale, NY) for 20 min in pulse mode (0.5 s pulses interrupted by 0.5 s breaks) at 40% of maximum power, resulting in a milky dispersion. Particle size was characterized using photon correlation spectroscopy (PCS) on a Malvern Zetasizer 3000 (Malvern Instruments, Malvern, UK) at 25 °C assuming a viscosity of pure water. Dispersions were stored at 25 °C for at least 2 days prior to further use; this enabled equilibration of lipid, F127, and water. Small-Angle X-ray Scattering Measurements (SAXS). SAXS measurements were performed on a Bruker Nanostar SAXS camera, with pinhole collimation for point focus geometry. The instrument source was a copper rotating anode (0.3 mm filament) operating at 45 kV and 110 mA, fitted with cross-coupled Go¨bel mirrors, resulting in Cu KR radiation, wavelength 1.54 Å. The SAXS camera was fitted with a Hi-star 2D detector (effective pixel size 100 µm). The optics and sample chamber were under vacuum to minimize air scatter. The sample-to-detector distance was chosen to be 650 mm, which provided a q-range of 0.08-0.320 Å-1. Samples were contained in 2 mm glass capillaries and temperature controlled by use of Peltier system accurate to (0.1 °C. Using SAXS, it was found that phase changes of the dispersions and bulk systems reach equilibrium after 20 and 60 min, respectively. Hence, samples of dispersions were allowed 30 min to equilibrate at temperature prior to measurement, while for bulk phases, 90 min of equilibration time was allowed. Scattering patterns were acquired over 30 min for bulk samples and 60 min for dispersions. Scattering files were background subtracted and normalized to sample transmission then integrated using Bruker AXS software v4.1.18 to the one-dimensional scattering function I(q), where q is the length of the scattering vector, defined by q ) (4π/λ)sin θ/2, λ being the wavelength and θ the scattering angle. The mean lattice parameter, a, is calculated from the interplanar distance, d, between two reflecting planes, given by d ) 2π/q. As the primary beam is point-focused instead of line-shaped, desmearing was not required for the L2 phase which shows only one broad peak. For the L2 phase, d is termed the characteristic distance. Crossed Polarized Light Miscroscopy (CPLM). Cross polarized light microscopy using a Zeiss Axiolab E microscope fitted with crossed polarizing filters and a magnification of 150× was used to observe the texture of the bulk mesophases. The textures were compared to those reported by Rosevear22 in order to assign the appropriate mesophases. A small amount of the bulk phase was placed on a microscope slide beneath a cover slip and, to minimize the effect of water loss during heating, the bulk phase was flooded with excess water. The sample was heated from room temperature to 80 °C at a rate of 1 °C min-1 using a Linkam HFS 91 heating stage and a TP-93 temperature programmer (Linkam, Surrey, England). Differential Scanning Calorimetry (DSC). DSC measurements were performed with a Perkin-Elmer DSC7 calorimeter at scanning (21) Geraghty, P. B.; Attwood, D.; Collett, J. H.; Dandiker, Y. Pharm. Res. 1996, 13, 1265-1271. (22) Rosevear, F. B. J. Soc. Cosmetic Chem. 1968, 19, 581-594.

9514 Langmuir, Vol. 22, No. 23, 2006

Dong et al.

Table 1. Comparison of Internal Lattice Parameters a (Å) and Phase Structure for GMO and Phytantriol Bulk and Dispersion (Disp) Systems with Addition of F127 in Excess Water at 25 °Ca

GMO

PHYT

[F127] in lipid (% w/w)

water content (% w/w)

form

space group

a (Å)

2.0 6.3 10.0

54.4 53.6 51.5 90.0

bulk bulk bulk disp

Pn3m Im3m Im3m Im3m

94.1 ( 0.6 128.3 ( 0.4 128.3 ( 0.4 129.7 ( 0.8

0.7 2.0 6.3 10.0

54.6 54.2 53.4 51.4 90.0

bulk bulk bulk bulk disp

Pn3m Pn3m Pn3m Pn3m Pn3m

68.2 ( 0.4 68.4 ( 0.3 68.3 ( 0.2 67.7 ( 0.2 68.6 ( 0.3

a Value for a and space group determined from individual SAXS I vs q plots.

rate of 5 °C min-1 under nitrogen. The samples (10-20 mg of bulk phases) were introduced into the aluminum pans, and an empty pan was used as a reference. Data acquisition and analysis were performed using software provided with the instrument.

Table 2. Effect of Lipid Type, Pluronic F127 Concentration, and Vitamin E Acetate (VitEA) Content on the Mean Particle Size and Polydispersity Index for Liquid Crystal Particle Dispersions Measured Using PCSa dispersion composition GMO + 10% F127 phytantriol + 5% F127 10% F127 20% F127 30% F127

mean particle size (nm)

polydispersity index

194.5 ( 0.1

0.148 ( 0.016

Effect of [F127] 270.3 ( 1.5 268.8 ( 0.9 259.8 ( 2.6 222.8 ( 0.5

0.233 ( 0.003 0.130 ( 0.019 0.209 ( 0.022 0.188 ( 0.013

Effect of [VitEA] at Constant 10% F127 phytantriol + 1% VitEA 297.8 ( 1.9 0.268 ( 0.006 10% VitEA 251.6 ( 0.7 0.237 ( 0.023 a Data are mean ( s.d. of three separate measurements at 25 °C. All dispersions contain either GMO (in the form of Myverol), phytantriol, or phytantriol + VitEA at 10% w/w of the total dispersion. VitEA concentrations expressed as mass VitEA/Pluronic F127 concentrations expressed as the percentage ratio 100(mass F127)/(total mass of lipids + mass F127).

Results Comparison of GMO and Phytantriol Bulk Phase Structure on Addition of Pluronic F127. SAXS patterns obtained for GMO and phytantriol with increasing amounts of water provided the phase progression from L2 phase to Ia3d to Pn3m cubic space groups with increasing water, in agreement with previous studies.17,23 CPLM revealed that the texture at the water/ phytantriol interface converted from isotropic to fully birefringent between 59 and 61 °C. The birefringence had the classical fanlike texture indicating the presence of HII phases as expected from the previous report but at higher temperatures than previously indicated.17 At approximately 65-66 °C, the HII phases transformed to isotropic, low-viscosity L2 phase. The effect of addition of F127 to the GMO-based mesophase in excess water is reported in Table 1. Addition of F127 at 2% w/w of the total GMO + F127 induced a transition in the mesophase structure from Pn3m to Im3m and an increase in lattice parameter from approximately 94 to 128 Å. Increasing the concentration of F127 up to 6.3% w/w did not induce any further increase in lattice parameter, in agreement with previous studies.10,11 In direct contrast, the incorporation of F127 into the mesophases formed by phytantriol at the same composition as those used for GMO had no effect on the observed space group or the measured lattice parameter. Structure and Stability of Sub-micrometer Dispersions of GMO and Phytantriol. The particle size of dispersions prepared during the course of the study were characterized by PCS. The mean particle sizes for the dispersions were all in the approximate range from 200 to 300 nm. The polydispersity indices were in the approximate range of 0.2-0.3, indicating a moderately heterogeneous system, likely indicative of the use of sonication as a preparative method compared to high-pressure homogenization. The GMO-based dispersion provided the lowest particle size of all of the dispersions investigated (194.5 ( 0.1 nm). Table 2 illustrates that there was little difference in particle size for the phytantriol-based dispersions with either increasing F127 concentration or with increasing vitamin E acetate content at constant F127 concentration. Phytantriol-based dispersions containing 9% w/w phytantriol and 1% w/w F127 (i.e., 10% w/w of total phytantriol + F127) were metastable for several months when stored at 25 °C. Some (23) Clogston, J.; Rathman, J.; Tomasko, D.; Walker, H.; Caffrey, M. Chem. Phys. Lipids, 2000, 107, 191-220.

Figure 2. I vs q plot for phytantriol dispersions at 25 °C (9% w/w in water) containing increasing concentration of F127. The concentration of F127 relative to phytantriol was (a) 10% w/w, (b) 20% w/w, and (c) 30% w/w. Note that the scattering profiles in Figure 2 are displaced on the vertical axis for clarity of presentation.

aggregation of lipid at the water/air interface was apparent but was easily redispersed with hand shaking. SAXS data taken 2 weeks after preparation revealed that the space group was Pn3m, as was the case for the bulk phase, and the lattice parameter was essentially identical at 68.6 ( 0.3 Å compared to 68.2 ( 0.4 Å for the bulk phase (Table 1). The SAXS I vs q plots for the dispersions obtained with the increasing F127 concentrations (Figure 2) revealed that there was no change in the phase structure or the lattice parameter even at F127 concentrations up to 30% relative to the phytantriol content (note that the scattering profiles in Figure 2 are displaced on the vertical axis for clarity of presentation). The change in space group for GMO bulk phase upon addition of F127 from Pn3m to Im3m described in the previous section was also observed in the GMO dispersions (Table 1). The phase structure of the GMO-based dispersion was identical to the equivalent bulk phase containing F127, with a space group of Im3m, and a lattice parameter of 129.7 ( 0.8 Å (see Table 1). GMO-based dispersions were visually more stable than the phytantriol-based system. Compared to the phytantriol dispersions, there was minimal aggregation of particles upon extended

Aqueous Phase BehaVior of Phytantriol

Figure 3. Effect of VitEA composition on bulk phytantriol phase behavior in excess water determined by CPLM and DSC. CPLM data as stacked bars: Diagonal stripe, isotropic/viscous; crosshatched, birefringent/viscous; vertical line, isotropic/fluid; DSC data: black circle, endothermic peak 1; white circle, endothermic peak 2.

storage. However, in contrast to the phytantriol-based dispersions, at extended times the GMO-based system exhibited a growing discoloration from milky white to creamy yellow, indicating the possibility of chemical changes in the sample such as lipid oxidation. However, SAXS results obtained after approximately

Langmuir, Vol. 22, No. 23, 2006 9515

3 months of storage did not reveal any changes in phase structure (data not shown). Phytantriol + Vitamin E Acetate MixturessEffect of VitEA Addition on Phytantriol Lyotropic Phase Structure and Transition Temperatures. Previous experience in this laboratory found that the addition of VitEA to phytantriol bulk phases suppressed the transition temperature of the QII + excess H2O to HII + excess H2O phase boundary.20 VitEA is structurally similar to phytantriol (Figure 1) and was used in this study to manipulate the phase structure exhibited by the phytantriol/water system. The effect of addition of increasing amounts of VitEA to phytantriol on the transition temperatures exhibited by the bulk phase obtained from CPLM data is illustrated in Figure 3. The isotropic/viscous, birefringent/viscous and isotropic/fluid regions were identified. SAXS data (discussed in the following paragraph) confirmed that the regions correspond to QII, HII, and L2 phases, respectively. A dramatic reduction in the QII + excess water to HII + excess water transition temperature was evident with increasing VitEA content; with a reduction from 60 °C to below 25 °C upon the addition of only 5% w/w VitEA to phytantriol. Temperatures for the phase transitions detected by DSC are overlaid in Figure 3. The DSC transitions coincide with the phase boundaries determined by CPLM, indicating that the two endotherms correlate with the structural changes that induce the changes in birefringence. The increasing VitEA concentration also suppressed the HII-to-L2 transition, but this phase transition

Figure 4. Panels A and C illustrate the phase structure for the bulk and dispersion systems, respectively, determined at the various temperatures with increasing VitEA concentration by SAXS where the symbols represent the following: Black circles, QII; white circles, QII + HII; black triangle, HII; white triangle, QII + HII + L2; black square, L2. Panels B and D are the partial (phytantriol + excess water) + VitEA phase diagrams determined from Panels A and C indicating the various phase regions. The range of coexisting QII + HII < 3% w/w VitEA were too narrow to be observed in this study. The dashed line indicates the likely HII-to-L2 phase boundary extrapolated from CPLM and DSC studies from the bulk phase system, as these composition/temperature combinations were not investigated explicitly using SAXS.

9516 Langmuir, Vol. 22, No. 23, 2006

Dong et al.

Figure 6. Temperature-dependence of the lattice parameters a [Å] (for QII (Pn3m), circles; HII, rectangle; and L2, triangular) for phytantriol with 2.7% w/w VitEA dispersions (90% w/w water), (full symbols); and the equivalent nondispersed bulk sample (55% w/w water), (open symbols). The individual I vs q plots from which this figure was derived are provided in the Supporting Information, Figure S1.

Figure 5. I vs q plot for bulk phytantriol containing 5% w/w VitEA at 20 °C. Indexing of Braggs peaks reveals a region of three-phase coexistence by the presence of peaks corresponding to two bicontinuous cubic (D and G) and reverse hexagonal (H). In this case, the D and G nomenclature is used in place of Ia3d and Pn3m space group nomenclature, respectively, for clarity in the figure. Note the overlap of some peaks resulted in broadening or splitting of apexes which likely indicates the presence of at least two unique reflections contributing to one peak.

is affected to a significantly lesser extent than the QII-to-HII transition. An expansion of the temperature range of coexistence for both the QII + HII and HII + L2 regions was also observed with increasing VitEA content, indicated by the coexistence of birefringent and isotropic textures in CPLM and the broadening of endotherms in DSC (see Supporting Information, Figure S2). The phases identified by SAXS and their location in the temperature-composition diagram of the phytantriol/VitEA/ water nondispersed system are shown in Figure 4, Panel A. The partial phase diagram shown in Figure 4, Panel B was constructed from combination of SAXS data from Panel A and CPLM/DSC data of equivalent composition from Figure 3. The SAXS data also revealed a reduction in the transition temperature for QII to HII from above 65 °C to between 15 and 30 °C with addition of 5% w/w VitEA, while for addition above 10% w/w, the transition temperature was suppressed to below 0 °C in agreement with CPLM and DSC. As was demonstrated by the CPLM and DSC data, the SAXS data showed that the magnitude of the suppression of the HII-to-L2 transition temperature was lower than that for the QII-to-HII transition. Expansion of the regions of coexisting QII and HII phases was again observed with increasing VitEA. At 5% w/w VitEA and 20 °C, a possible region of three phase coexistence was also observed in these systems. Peaks shown in the I vs q plot in Figure 5 were indexed to HII, Pn3m, and Ia3d cubic phases. The stability of VitEA + phytantriol dispersions over time was similar to that of the pure phytantriol dispersions. The temperature-composition diagram composed using the SAXS data in Figure 4, Panel C and the simplified partial phase diagram in Panel D illustrate that the internal structure of the dispersed particles closely follows that of the equivalent parent bulk phases

from Panels A and B. The addition of F127 to prepare the dispersions with VitEA had minimal effect on the phase structure. This was demonstrated by the indistinguishable lattice spacings (Figure 6), which was also the case for the phytantriol dispersions (in Figure 2). Slight differences were observed between the QIIto-HII conversion temperatures for bulk and dispersed systems. Addition of 2.7% w/w VitEA to phytantriol resulted in the QIIto-HII conversion occurring at 35-50 °C for the bulk phase compared to 45-50 °C for the equivalent dispersed system (see Supporting Information, Figure S1). The HII-to-L2 phase conversion also generally occurred at lower temperatures for the dispersed systems than for the bulk phases. For both pure phytantriol and phytantriol with 1.0% w/w VitEA, there was no evidence of formation of HII phase within the resolution of the methods used in this study, as the structure of the particles transformed directly from QII at 60 °C to L2 at 65 °C.

Discussion Manipulation of Phytantriol Lyotropic Phase Behavior. In a previous study on the lyotropic phase behavior of phytantriol,17 the QII-to-HII and HII-to-L2 transitions occurred at significantly lower temperatures than those determined in this study; in particular the QII-to-HII transition was reported to occur at approximately 40 °C compared to >55 °C in this study. Hato et al. (2004) have investigated the lyotropic phase behavior of 1-O-phytanyl-β-D-xyloside, which has same the phytanyl lipid chain as phytantriol, which also showed similar lyotropic behavior to phytantriol in this study.24 In contrast, phytanyl glycerate displays a HII phase at ambient temperatures in excess water despite having the same phytanyl chain as 1-O-phytanyl-β-Dxyloside and phytantriol, demonstrating the importance of polar headgroup structure in the mesophase formation.25 The phytantriol bulk phase in water exhibits similar thermal behavior to GMO26 and MLO15 where the lattice parameters of all mesophases decreased with increasing temperature. This is understood to be due to increased disorder in the hydrocarbon (24) Hato, M.; Yamashita, I.; Kato, T. Abe, Y. Langmuir 2004, 20, 1136611373. (25) Boyd, B. J.; Whittaker, D. V.; Khoo, S. M.; Davey, G. Int. J. Pharm. 2006, 309, 218-226. (26) Qiu, H.; Caffrey, M. Biomaterials, 2000, 21, 223-234.

Aqueous Phase BehaVior of Phytantriol

chains increasing the curvature of the lipid bilayers causing a reduction of the unit cell15 (see Figure 6). For phytantriol, SAXS data indicated that the QII-to-HII conversion occurred at 65-70 °C, rather than 60 °C indicated by CPLM or 57 °C by DSC. The HII-to-L2 conversion temperatures determined by SAXS were also higher in comparison to those from CPLM or DSC. This slight discrepancy was observed in all bulk systems tested with exception of those with high VitEA content (>5% w/w). This discrepancy contradicts Abe and Takahashi who observed excellent agreement between SAXS and DSC on GMO-water systems.27 The transition between phases observed in CPLM took place within seconds of temperature increases, indicating that the transition temperature is likely to be accurate in the CPLM investigations. The 90 min equilibration time used in SAXS after changes in temperature before the scattering patterns were acquired was intended to allow the attainment of thermal equilibrium; however, the possibility that the slight discrepancies between DSC/CPLM and SAXS are due to a combination of experimental uncertainty and differences in thermal conductance and equilibrium of samples cannot be discounted. Simultaneous SAXS and DSC will be conducted in the future to further investigate this discrepancy; however, in the broad context of this paper, the differences do not impact on the important conclusions drawn from these studies. The formation of QII and HII phases are the result of the interplay between local and global constraints in the system.5 The former impacts on the local interfacial curvature and is dictated by the effective shape of the amphiphilic molecules while the global constraints depend on the composition of the mixture. The effect of molecular shape on interfacial curvature has been described in terms of molecular packing, the so-called critical packing parameter (CPP).28 It should be noted that the phytantriol and vitamin E acetate utilized in these studies are nominally 95% and 96% pure, respectively, and consequently, the effect of impurities on phase structure and phase coexistence may impact on the observed phase behavior presented in this report, a complication acknowledged in an analogous study on the monolinolein-water system.15 Vitamin E acetate, with its chromanol group in place of more hydrophilic hydroxyl groups of phytantriol, will qualitatively increase the volume of hydrophobic portion of the molecule, dictating that VitEA would likely reside in the hydrophobic regions of the lipid bilayer imparting a negative curvature on the system. The consequent transition from QII to HII with increasing VitEA concentration is analogous to addition of diolein to GMO.13 On heating, nonionic surfactants will often exhibit two melting points, that correlate with crystalline-to-liquid crystalline and liquid crystalline-to-isotropic melt transitions on CPLM. The lower-temperature endotherm is attributed to chain melting and the higher-temperature transition to intermolecular hydrogen bonding.29,30 DSC data revealed two endotherms for the phytantriol-based lyotropic systems (Figure 3) which correlated directly with phase changes observed by CPLM. The lower temperature transition was observed to be strongly dependent on VitEA concentration, indicating that it is predominantly a chain packing transition, while the second, higher temperature transition was relatively insensitive to VitEA concentration indicating that a rearrangement in headgroup region is more likely for this transition. These observations indicate an analogy between the (27) Abe, S.; Takahashi, H. J. Appl. Crystallog. 2003, 36, 515-519. (28) Israelachvili, J. N.; Mitchell, D. J.; Ninham, B. W. J. Chem. Soc., Faraday Trans. ΙΙ 1976, 72, 1525-1568. (29) Jeffrey, G. A.; Bhattacharjee, S. Carbohydr. Res. 1983, 115, 53-58. (30) Boyd, B. J.; Drummond, C. J.; Krodkiewska, I.; Grieser, F. Langmuir 2000, 16, 7359-7367.

Langmuir, Vol. 22, No. 23, 2006 9517

behavior in thermotropic liquid crystalline systems and the lyotropic systems studied here. A progressive broadening of the temperature range for the coexistence of QII + HII with increasing VitEA content was evident in CPLM (coexistence of birefringent and isotropic textures represented by gaps between the segments of columns in Figure 3), DSC (endotherm broadening in Supporting Information, Figure S2), and SAXS data. The suggestion of de Campo et al. that the coexistence for QII and HII in monolinolein is indicative of impurities is consistent with the findings in this study if we consider VitEA to be acting as a minor impurity in this system.15 To our knowledge, this is the first time that Ia3d cubic phase has been observed during the Pn3m-to-HII transition; however, the Im3m cubic phase has been observed during Pn3m-to-HII transition using lauric acid and dilauroylphosphatidylcholine.31 However, another study using 2:1 (mol/mol) C12 saturated fatty acid and phosphatidylcholine mixtures in excess water showed a phase transformation from Im3m to Pn3m/HII with increasing temperature, which is analogous to our Pn3m-to-Ia3d/HII transition.32 This observation may be understood in the context of the ‘breathing mode’ concept introduced by de Campo et al.15 to describe the efflux/influx of water on heating/cooling of mesophases. As the Ia3d space group exists at a lower water content than Pn3m in the phytantriol/water phase progression at ambient temperature, heating induces a similar effect as dehydration of internal bulk phases leading to formation of the Ia3d space group.15 This could be an important finding if the release of incorporated active species from these matrixes depends on the structural space group. Characterization of Phytantriol-Based Cubosomes and Hexosomes. For phytantriol dispersions, the stability of the Pn3m cubic phase to the addition of large concentrations of F127 was originally hypothesised to be due to the small lattice spacing of the phytantriol-water Pn3m space group (∼68 Å) inhibiting the partition of F127 into the internal cubic nanostructure. However, this is unlikely as monolinolein displays a similar resistance to phase change with addition of F127, despite displaying a lattice spacing similar to GMO.15 The change of space group from Pn3m to Im3m is likely due to the insertion of hydrophobic polypropylene block of F127 into the bilayer of GMO cubic phase. Hence, an alternative explanation for the phase stability may be related to the mode of interaction between F127 and the particle surfacesa lesser affinity of the PPO to partition into the phytantriol bilayer may lead to the polymer simply adsorbing at the surface, thereby not influencing the lattice parameter or space group. The branching of methyl groups on the hydrocarbon chain for phytantriol may provide an unfavorable environment for the insertion of F127 into the internal bilayers. In support of this hypothesis, the same Pn3m-to-Im3m conversion as observed for phytantriol when using vitamin E TPGS as the stabilizer;19 the hydrophobic portion of vitamin E TPGS is similar to phytantriol in a structural sense and hence would be expected to interact more strongly with the phytantriol bilayer than F127 facilitating the structural conversion. Similar to monolinolein, we also found that the conversion from the HII to L2 phase was at a higher temperature for the nondispersed bulk phase than that of the dispersed systems (Figure 6). However, in the previous study, an explanation for the differences seen in the monolinolein system was not provided.15 (31) Squires, A.; Templer, R. H.; Ces, O.; Gabke, A.; Woenckhaus, J.; Seddon, J. M.; Winter, R. Langmuir 2000, 16, 3578-3582. (32) Templer, R. H.; Seddon, J. M.; Warrender, N. A.; Syrykh, A.; Huang, Z.; Winter, R.; Erbes, J. J. Phys. Chem. B 1998, 102, 7251-7261.

9518 Langmuir, Vol. 22, No. 23, 2006

It is also not clear from these studies why the difference should exist. In light of the high surface area and small size of the particles relative to the bulk samples, it is likely that the effect is either due to the presence of F127 in the dispersed form or due to differences in phase structure in the interfacial regions of the materials. The presence of F127 in the dispersed system may in part be responsible for the differences in thermal behavior; however, no changes in lattice parameter between the bulk and dispersed systems is apparent, and it is not possible to suggest a mechanism based on F127 alone from the data at hand. A second possibility may be the relatively small number of unit cells comprising the sub-micrometer particles imparts significant constraints on the local structure in the dispersed particles that is not dominant in the bulk systems. This is also likely in consideration of the requirements for altered packing at the extreme interface of the liquid crystalline structures with excess water in order to minimize the contact of the hydrophobic chains with water; however, the exact nature of the arrangement of the amphiphilic molecules at this interface is still not known, but it is reasonable to assume that the effect is much greater for the sub-micrometer dispersions in light of the much greater surface area of liquid crystal in contact with bulk solution.

Conclusions In this paper we reported the manipulation of phytantriolwater bulk liquid crystalline and dispersed liquid crystalline behavior with addition of small amount of Vitamin E acetate. The main conclusions are that

Dong et al.

•Small amounts of VitEA added to the phytantriol-water system suppresses the temperature of the QII-to-HII-to-L2 phase transitions. •Nanostructured particles (cubosomes and hexosomes) with versatile thermal behavior can be prepared by dispersion of phytantriol/VitEA mixtures in the presence of Pluronic F127 and retain the internal structure of the parent bulk phase. •In contrast to GMO-based cubosomes, increasing the F127 concentration does not result in any change to the lattice parameter or space group of the phytantriol base cubosomes. This suggests a different mode of interaction between the particle and the stabilizer. The ability to tailor the nanostructure of these systems will be utilized in future studies to enable the determination of the adsorption of cubosomes and hexosomes at surfaces and as a means to control loading and release of actives in drug delivery applications. Acknowledgment. The authors thank the Australian Institute of Nuclear Science and Engineering for funding this project under Grant No. AINGRA05185. Yao-Da Dong thanks Monash University for support in the form of an Australian Postgraduate Award scholarship. Supporting Information Available: I vs q plots and DSC curves. This material is available free of charge via the Internet at http:// pubs.acs.org. LA061706V