High-Throughput Screening of Saturated Fatty Acid Influence on

Mar 29, 2016 - ... Engineering and Health, RMIT University, Melbourne, Victoria 3000 Australia ... Phone: +61 3 9545 2630., *E-mail: calum.drummond@rm...
1 downloads 0 Views 9MB Size
Article pubs.acs.org/Langmuir

High-Throughput Screening of Saturated Fatty Acid Influence on Nanostructure of Lyotropic Liquid Crystalline Lipid Nanoparticles Nhiem Tran,†,‡,∥ Adrian M. Hawley,‡ Jiali Zhai,† Benjamin W. Muir,† Celesta Fong,† Calum J. Drummond,*,†,§ and Xavier Mulet*,† †

CSIRO Manufacturing, Clayton, Victoria 3168 Australia SAXS/WAXS Beamline, Australian Synchrotron, Clayton, Victoria 3168 Australia § School of Applied Sciences, College of Science, Engineering and Health, RMIT University, Melbourne, Victoria 3000 Australia ‡

S Supporting Information *

ABSTRACT: Self-assembled lyotropic liquid crystalline lipid nanoparticles have been developed for a wide range of biomedical applications with an emerging focus for use as delivery vehicles for drugs, genes, and in vivo imaging agents. In this study, we report the generation of lipid nanoparticle libraries with information regarding mesophase and lattice parameter, which can aid the selection of formulation for a particular enduse application. In this study we elucidate the phase composition parameters that influence the internal structure of lipid nanoparticles produced from monoolein, monopalmitolein and phytantriol incorporating a variety of saturated fatty acids (FA) with different chain lengths at varying concentrations and temperatures. The material libraries were established using high throughput formulation and screening techniques, including synchrotron small-angle X-ray scattering. The results demonstrate the rich polymorphism of lipid nanoparticles with nonlamellar mesophases in the presence of saturated FAs. The inclusion of saturated FAs within the lipid nanoparticles promotes a gradual phase transition at all temperatures studied toward structures with higher negative surface curvatures (e.g., from inverse bicontinuous cubic phase to hexagonal phase and then emulsified microemulsion). The three partial phase diagrams produced are discussed in terms of the influence of FA chain length and concentration on nanoparticle internal mesophase structure and lattice parameters. The study also highlights a compositionally dependent coexistence of multiple mesophases, which may indicate the presence of multicompartment nanoparticles containing cubic/cubic and cubic/hexagonal mesophases.

1. INTRODUCTION The development of improved controlled release methods and molecular targeting of disease states using therapeutic soft nanoarchitectures derived from self-assembled amphiphiles is fundamental to the application of nanomedicine. Soft colloidal nanostructures such as hexosomes and cubosomes, which are stable lyotropic liquid crystalline nanoparticles with inverse hexagonal and inverse bicontinuous cubic phases, have been investigated as drug delivery vehicles for more than two decades and a wide variety of bioactives have been incorporated into both phases.1−10 Low energy transformations between thermodymically stable two- and three-dimensional nanostructures or between polymorphic cubic mesophases have been used for this purpose. The interconversion between nanostructures arises due to an interplay between the hydrophobic effect and geometric packing constraints of the molecule. The resultant structures that form may be manipulated by various global constraints such as temperature, pH, pressure, hydration, composition, and ionic strength to achieve nanoarchitectures with different rheological and topological properties.11 An evolving strategy is the compositional variation of monoacylglyceride (MAG)- and © XXXX American Chemical Society

phytantriol-lipid mixtures with fatty acids (FAs), hydrocarbons, glycolipids, and phospholipids, to engineer desirable properties such as drug diffusion control, size selectivity and stimuli responsiveness.12−18 For instance, Negrini et al. have used monolinolein and sucrose stearate bicontinuous cubic phases (bcps) as filtering membranes with a tunable molecular cutoff;6 with Thapa and co-workers improving sustained release of tacrolimus from monoolein systems prepared with short chained FAs (C12 and C14).19 In addition to this the pH triggered release of excipients has been achieved through the incorporation of unsaturated linoleic acid in monolinolein.20 Many of these nonlamellar architectures are present in biomembranes, for which parallel structures have been observed. These soft nanoarchitectures have been employed as model membranes and local structural perturbations at the interface, arising in response to external stimuli such as the uptake of guest molecules, have received considerable attention. For example, the lamellar-to-cubic lipid membrane phase transition is Received: October 13, 2015 Revised: March 11, 2016

A

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 1. Chemical structure of monoolein (MO), monopalmitolein (MPO), phytantriol (PHYT), and Pluronic F127.

Figure 2. Cryo-TEM of monoolein nanoparticles doped with decanoic acid. Images show typical cubosomes (a), hexosomes (b), and emulsified microemulsions (c) formed at increasing concentration of decanoic acid (capric acid, C10) in the lipid mixture. rMO is 0.13, 0.52, and 2.07 in (a), (b), and (c), respectively.

important for engineering “fit-for-function” materials. Recent advances in automation and combinatorial methods can streamline materials development workflows, from synthesis, characterization, formulation, and the assessment of end-use functional performance, to accelerate materials development.24,25 In this study, an automated combinatorial highthroughput approach has been used to formulate and characterize lipid-FA nanoparticle libraries that is aimed at expanding our current knowledge of the structure−property relationships in these systems. We have performed a systematic study of the influence of saturated FA (C7 to C16) incorporation upon the resulting nanoarchitecture of three well characterized, nonlamellar phase forming lipids, namely, monoolein (MO), monopalmitolein (MPO), and phytantriol (PHYT) (Figure 1). Partial phase diagrams of the MO-FA, the MPO-FA, and the PHYT-FA nanoparticulate systems dispersed by block copolymer Pluronic F127 in excess water have been obtained by high throughput synchrotron small-angle X-ray scattering (SAXS) and are reported herein.

important in the context of cellular membrane functions.21 Recent work by Angelova et al. has captured long-lived intermediate states of the bilayer membrane vesicle to cubic transition upon protein trafficking at the lipid/water interface, wherein multicompartment architectures have been identified.22 The reproducible programmable design and fine control of nanoscale structure is a major challenge that must be overcome for future innovations in the field. Next generation soft colloidal nanostructures are responding to the demands of personalized medicine through the integration of excipients for multiplatform capabilities in therapeutics and diagnostics (e.g., imaging). The desire for smarter nanomaterials is driving research in advanced nanosystems and theranostics with controllable multifunctional properties. While there is a wealth of knowledge pertaining to the reproducible preparation of nanoparticles with isotropic properties, for applications in theranostics, anisotropy in shape, nanostructure and/or surface properties offer a route to multiplicity in functionality. An interesting development in this area is a recent observation by us that discrete lipid nanoparticles can form with biphasic cubic (QII P)−cubic(QII D) and cubic(QII D )−hexagonal (HII) nanostructures,22,23 in a capric acid (C10)− monoolein system.23 In other words, the creation of Janus nanoparticles by lipid self-assembly.23 Hence, with the burgeoning of potential drug candidates and therapies, the ability to deconvolute the relationship between structure, composition, and function is becoming increasingly

2. EXPERIMENTAL SECTION 2.1. Materials. Monoolein (MO) and monopalmitolein (MPO) were obtained from Nu-chek-Prep, Inc. (Elysian, MN) with purity > 99% determined by using high-performance liquid chromatography. Phytantriol (PHYT) was a gift from DSM Nutritional Products. Saturated FAs with carbon chain length from C7 to C16 were purchased B

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 3. Partial phase diagram of the MO-FA nanoparticle libraries at 25 °C (A). Each individually prepared nanoparticle formulation with fatty acid chain length from C7 to C16 is displayed on the y-axis with the molar ratio of fatty acid to MO (mol:mol), rMO, given along the x-axis. Legend: EME (emulsified microemulsion), HII (hexagonal phase), QIID (diamond cubic phase, space group Pn3m), QIIP (primitive cubic phase, space group Im3m), Lc (lamellar crystal). Representative SAXS images of nanoparticle formulation of MO (B1), MPO (B2), and PHYT (B3) added with palmitic acid (C16) show the presence of lamellar crystals (Lc) at 25 °C. The molar ratios of palmitic acid to lipids are rMO = 0.35, rMPO = 0.32, and rPHYT = 0.32. form Sigma-Aldrich (St. Louis, MO). Pluronic F127 and ethanol were also purchased from Sigma-Aldrich. 2.2. Preparation of Lipid Lyotropic Liquid Crystalline Nanoparticles. Traditionally, lyoptropic liquid crystalline nanoparticle dispersions are made one at a time by mixing the amphiphile and payload, then adding an aqueous stabilizer solution and dispersing through ultrasonication or homogenization. This time-consuming process has been superseded by a robotic handling process in the current study to produce three libraries with a gradient in composition of FA concentration in the lipid matrix. A combinatorial library of 120 nanoparticle materials was generated for each type of the three bulk lipids, that is, MO, MPO, and PHYT, using the automated system (Chemspeed, Switzerland) at the CSIRO Rapid Automated Materials & Processing Centre (RAMP). Increasing amounts of FAs with carbon chain lengths from C7 to C16 were added to the bulk lipids to achieve a wide range of lyotropic liquid crystalline nanostructures. The ratio of FA to lipid (mol:mol) was used to define the material composition and given as r=

Australia). Pluronic F127 in Milli-Q water (18.2 MΩ) was added to the dried mixtures at 10 wt % to the total amount of bulk lipids and FAs. Mixtures were then sonicated by an automated probe sonicator on a Chemspeed platform for 8 min per MO sample and 10 min per PHYT sample.24 After sonication, the plate was sealed and the resultant material libraries were kept at room temperature for further examination. The presence of homogeneous opaque dispersions confirmed the effectiveness of the steric stabilizer. Figure 2 shows a cryo-TEM image of a representative nanodispersion of the MO-FA (C10) system illustrating the characteristic structures of the cubosomes, hexosomes and water-in-oil microemulsion (emulsified microemulsions)26 formed with increasing concentration of FA. 2.3. Synchrotron SAXS Experiments. SAXS experiments were performed at the SAXS/wide-angle X-ray scattering (WAXS) beamline at the Australian Synchrotron. The instrument used an X-ray of wavelength of λ = 1.128 Å (11.0 keV) with a typical flux of approximately 1013 photons/s. The sample to detector distance was chosen as 1.6 m which provided a q-range of 0.01−0.5 Å−1 (scattering vector q = 4π sin(θ)/λ where θ is the scattering angle and λ is the wavelength). Two-dimensional X-ray diffraction images were recorded on a DecrisPilatus 1-M detector using in-house IDL-based ScatterBrain software.27 The scattering images were integrated into one-dimensional plots of intensity versus q for phase identification. A silver behenate standard (d = 58.38 Å) was used for calibration. The exposure time for each sample was 1 s. The library of prepared materials (100 μL) were loaded into a 96-well plate format that was directly compatible with the high throughput sample holder at the SAXS-WAXS beamline at the Australian Synchrotron. Movement of the plate in relation to the beam was

nfatty‐acid nlipid

where nfatty‑acid and nlipid are the number of mole of saturated fatty acid and bulk lipid (MO, MPO, PHYT) respectively. We estimate that potential error in compositional makeup to be within ±5%. To prepare the libraries, MO, MPO, PHYT, and FA stock solutions were dissolved in absolute ethanol separately. Mixtures of lipids and FAs were added to each well of a 96-well deep well (2 mL volume) block (Greiner Bio-One, Interpath Inc., Australia). Ethanol was then evaporated overnight using a centrifugal evaporator (GeneVac, NSW, C

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

volume ν, which also leads to higher CPP and more inverse structures. Simplistically, MO may be considered a wedge shaped molecule, with a relatively small headgroup with respect to the footprint of its hydrophobic tail that has inherent chain splay derived from cis unsaturation of the glycerol backbone; as such, it has a preference for type II or inverse phases.38 FAs in water are capable of forming self-assembly structures dependent on pH.39,40 They are, however, expected to partition preferentially into the MO bilayer, to favor the formation of mesophases with increasing negative interfacial curvature. Hence, with increasing FA chain length there is a decrease in the concentration required to drive the CPP to a more negative interfacial curvature. Specifically for the C7 FA, the transition from QIIP → QIID occurred at around rMO = 0.68 and this compares to rMO = 0.11 for C16 FAs (Figure 3A). Consistent with this, is the presence or broadening of the HII phase that was not observed in the heptanoic acid (C7) series. There is a concomitant reduction in the extent of the QIIP domain as a function of FA chain length. At longer FA chain lengths, packing frustration must also be considered. The concept of packing frustration hinges upon the idea of the optimized tiling (packing) of elements of uniform geometry, for example, cylinders into hexagonally close packed symmetries (hcp).34 Simple geometric considerations of the hcp of cylindrical elements, reveals that such organization generates “void” space that is energetically unfavorable and is compensated by being filled by amphiphile alkyl chains that must extend beyond their preferred conformation to fill the interstices. Relief of the packing frustration of the chains may be afforded by partitioning of added alkanes into the “void” volume (Figure S5).41−44 FAs with longer acyl chains are predicted to fill in the void space more effectively, thus stabilizing the inverse hexagonal phase. This manifests as extended HII regions in the phase diagram (C8−C14). The addition of FA may also lead to nonuniform distribution of lipids around the cylinders of the HII phase with longer chain molecules arranged so that the “void” volume is filled. The HII region of the phase diagram exists over a broader and higher temperature range, with the EME domain enhanced at low FA chain lengths (C7−C11) (Figures S1−S4). Packing frustration can also be an important factor in cubic phases. Shearman et al. have also demonstrated that differences in the relative phase behavior of the three bicontinuous cubic phases under the constraint that their interfaces have constant mean curvature, can be understood in terms of subtle differences in packing frustration.42 These authors have found that the gyroid cubic has the least packing stress, and at low water volume fraction, the primitive cubic has the greatest packing stress. This preference for greater negative interfacial curvature is consistent with an increase in hydrophobic lateral packing and splay as a result of greater chain mobility with temperature. Lamellar crystals (Lc) are present at high concentrations with the longer chain length FAs (C13−C16). At 25 °C, most of the FAs are still in solid crystalline form with their melting temperatures given in the Supporting Information, Table S1. However, scattering of a crystalline phase was only observed at high concentrations with longer chain FAs. This suggests that MO helped solubilize shorter chain FAs at all concentrations and longer chain FAs at low concentrations. Once the saturation has been reached, longer FAs remained as lamellar crystalline particles suspended in solution. Mixed phases of Lc + QIID and Lc + HII phases were observed and are attributed to solubility saturation of FAs and phase separation from the lipid. This is also evident from the fact that the repeat distances of the lamellar

automatically controlled by motors via a custom-designed program. The combination of the automated stage and the short exposure time requires no more than 7 min to scan 96 materials. The sample temperature was controlled by a circulating water bath for experiments at 25, 37 , 40, 45, and 50 °C. 2.4. SAXS Data Analysis. The one-dimensional SAXS data were analyzed using an in-house automated software, RapidPhaseIdent. This program identifies lyotropic liquid crystalline phases and calculates the lattice parameter of each lyotropic liquid crystalline phase. Phase identification was based on the relative distance of the Bragg peaks in the scattering profile, which corresponds to diffraction planes defined by their (hkl) Miller indices. Lattice parameter a was calculated using the equation a = d(h2 + k2 + l2)1/2 for cubic phase or a = d(h2 + k2 + hk)1/2 for hexagonal phase where d is the spacing between the diffraction planes, defined by Bragg’s law d = 2π/q. Additionally, an IDL-based AXcess software package was used to manually analyze samples exhibiting scattering profiles of multiple lyotropic liquid crystalline phases.28

3. RESULTS AND DISCUSSION 3.1. Phase Behavior of the MO-FA Library. MO has been the amphiphile of choice for many years for the formulation of cubic based nanoparticle dispersions,3,5 and its phase behavior is well-known.29−31 FAs are endogenous membrane components known to modulate membrane structure and impact biological function. As such, they are commonly used as adjuvants in cosmetics and pharmaceuticals for transdermal applications.32 Figure 3A presents the partial phase diagram for the Pluronic F127-dispersed MO-FA systems at 25 °C. The phase diagrams at T = 37, 40, 45, and 50 °C are provided in the Supporting Information (Figures S1−S4, S14). In the absence of FA, the colloidal dispersions have a QIIP nanostructure which is consistent with previous literature reports.15 In general, the parent primitive cubic (rMO = 0) responded to increasing amounts of FA by undergoing a phase transformation from QIIP→ QIID →HII → EME (emulsified microemulsion). Representative SAXS scattering profiles are present in Figure 3B. This phase sequence is similar to that of previously reported MO−oleic acid and MO−capric acid systems.14,15,17 However, unlike the MO−oleic system, no inverse discontinuous micellar cubic (Fd3m) phase was observed in the current study. The QIIP and QIID phases may be described as a single, continuous bilayer draped over a surface which subdivides space into two noninterpenetrating, congruent water networks.33−36 The dimensionless critical packing parameter (CPP),37 has provided an extremely useful measure of aggregation topology. This parameter can be approximated from estimates of amphiphile dimensions, such that CPP = ν/aHL, where ν is the hydrophobic tail volume, aH is the headgroup area, and L is the hydrophobic chain length. CPP characterizes an average, local packing constraint and permits amphiphiles to be categorized by their “shape”. From such geometric constraints and in the absence of interaggregate interactions, the preferred direction of curvature for inverse structures (type II) such as the bcp and inverse hexagonal phases observed in the current study, CPP > 1; this compares to CPP = 1 for lamellar phases and CPP < 1 for normal or type I structures. The influence of temperature and solubilized hydrophobic compounds to the phase behavior of lyotropic liquid crystalline materials can be estimated by examining the CPP. For example, in general, increasing the temperature will result in more disorder in the hydrophobic chain, leading to higher CPP and the formation of more curved inverse structures such as inverse bicontinuous cubic phase and inverse hexagonal phase. Similarly, the addition of hydrophobic materials such as oils or fatty acids will increase the effective tail D

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 4. Lattice parameters of the MO-FA nanoparticles at 25 °C as a function of FA to MO molar ratio, rMO. The FA chain length varies from C7 to C16. The symbols for each mesophase are as follows: QIIP (blue tilted square), QIID (green plus sign), HII (yellow up triangle), EME (red circle), Lc (pink square). Note: for EME, a nonordered L2 phase, the value reported is measured from the single broad peak in the scattering profile, representing the average micelle separation (characteristic distance) of the L2 phase.

per milliliter (Supporting Information, Table S1), while the concentrations of FAs in nanoparticles are in a much higher range of milligram per milliliter, suggesting Lc is the lamellar

phase were independent of temperature and concentration, which will be discussed in the following sections. Additionally, the solubility of these FAs in water is in the range of microgram E

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 5. Partial phase diagram of the MPO-FA nanoparticle libraries at 25 °C. Each individually prepared nanoparticle formulation with fatty acid chain length from C7 to C16 is displayed on the y-axis with the ratio of fatty acid to MPO (mol:mol), rMPO, given along the x-axis. Legend: EME (emulsified microemulsion), HII (hexagonal phase), QIID (diamond cubic phase, space group Pn3m), QIIP (primitive cubic phase, space group Im3m), Lc (lamellar crystal).

MO, with a slightly shorter carbon backbone, C16 versus C18. As for its homologue, it bears cis unsaturation at the C9 position of its hydrophobic tail (C16:1c9); as such its shape parameter is also conical and thus it also preferences type II lyotropic liquid crystalline mesophases. It is not unexpected therefore, that its phase behavior in the presence of FA is qualitatively very similar to MO, with the same phase transition sequence observed with increasing FA concentration and molecular weight, namely, QIIP→ QIID → HII → EME (Figure 5). In general, however, QIIP and QIID occupy a more extended region of the compositional space; with the phase transition QIIP → QIID occurring at higher FA concentration compared to MO (Figure 5). As an example, for the MO-C9 FA system, the QIID + QIIP was observed at rMO = 0.18, that compares to rMPO = 0.23 (Figure 5). Additionally, the HII phase region in the phase diagram of MPO-FA system is much smaller compared to that of its C18:1c9 counterpart (Figure 5). The results suggest that MO is more sensitive to the addition of FA’s than MPO which is attributed to molecular structural differences between the two lipids. The shorter chain results in a more cylindrical shape; therefore, the cubic phases in MPO are more stable (bicontinuous cubic phases have lower negative interfacial curvature than HII phase). The lamellar crystal phase (Lc) was present as a mixed phase with HII and QIID with FA C12−C15, as well as with QIIP and QIID with FA C16, suggesting inadequate incorporation (phase separation) of these longer chained fatty acids which was partially resolved at higher temperatures. Representative scattering profiles of MPO nanoparticles with the Lc phase is provided in Figure 3B. Phase diagrams of the MPO-FA nanoparticle library at 37, 40, 45, and 50 °C are provided in Supporting Information Figures S6−S9 and S15. The LPs of MPO-FA system are presented in Figure 6. Similar to the MO-FA system, the addition of increasing amounts of FA resulted in a gradual reduction of the LP of all mesophases except the Lc. For instance, for MPO-C9 FA, as rMPO increased, LP of the QIIP phase lowered from 158 to115 Å (rMPO = 0.23) at which point a phase transition to the QIID phase occurred (LP = 87 Å). The LP gradually dropped to 72 Å at rMPO = 0.52 when the QIID

phase observed in the experiments. The elevated temperature appeared to help the solubility of FAs in MO. As the temperature increased from 25 to 50 °C, the intensity of the scattering peaks attributed to the Lc phase gradually decreased, with the Lc phase only observed in MO samples with high concentration of C16 FA added at 50 °C. Representative SAXS scattering profiles of MOFA with Lc phase are shown in Figure 3B. The lattice parameter (LP) of each phase in the MO-FA partial phase diagram was calculated from the scattering data using the relative position of the scattering peaks. The LP’s of the liquid crystalline dispersions with respect to rMO for each FA are presented in Figure 4. It should be noted that for an EME, which is a nonordered L2 phase, the value reported is measured from the single broad peak in the scattering profile, representing the average micelle separation (characteristic distance) of the L2 phase. Without FA, MO cubosomes stabilized with F127 have a LP of about 150 Å, consistent with previous reports.17 In general, the LPs decrease gradually as a function of FA incorporation in the MO bilayer, with phase transitions identified by an abrupt discontinuity. Hence the LPs of the phases decreased in the order QIIP (150−120 Å) > QIID (90−70 Å) > HII (61−47 Å). Previous studies reported a similar decrease in LP of the lyotropic liquid crystalline phases with an increased concentration of a solubilized hydrophobic additive.26 At high FA chain length the Lc phase was also observed, usually as a mixed phase system (e.g., Lc + HII for C13−C15 samples; and Lc + QIID for C16). The fact that both the d-spacing of the lamellar crystal and the LP of accompanying phase were invariant with FA and water concentrations suggests phase separation and incomplete solubility of the FA in MO. The lamellar repeat distances (dspacing) of the Lc phase are presented in Table S2. It is also notable that the d-spacings of Lc phases linearly increased with FA chain length. The observed values of Lc d-spacing were similar to previously reported d-spacings for anhydrous crystalline evennumbered chain fatty acids.39 3.2. Phase Behavior of MPO Nanoparticles in the Presence of Fatty Acids. MPO is structurally closely related to F

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 6. Lattice parameters of the MPO-FA nanoparticles at 25 °C as a function of FA to MPO molar ratio, rMPO. The FA chain length varies from C7 to C16. The symbols for each mesophase are as follows: QIIP (blue tilted square), QIID (green plus sign), HII (yellow up triangle), EME (red circle), Lc (pink square). Note: for EME, a nonordered L2 phase, the value reported is measured from the single broad peak in the scattering profile, representing the average micelle separation (characteristic distance) of the L2 phase.

to HII transition happened (Figure 6). Similar to the case of MO, the lamellar repeat distances of the Lc phase were invariant of FA concentration, rMPO, for the C12−C16 FA-MPO systems. The dspacings of Lc phase measured in MPO samples were nearly identical with those of MO samples and similar to the reported d-

spacing values of fatty acid solid crystals (Table S2). This result further confirmed the phase separation of FA and MPO and that the source of Lc scattering was solid fatty acid crystals. 3.3. Phase Behavior of PHYT Nanoparticles in the Presence of Fatty Acids. The phase diagram of PHYT-FA G

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 7. Partial phase diagram of the PHYT-FA nanoparticle libraries at 25 °C. Each individually prepared nanoparticle formulation with fatty acid chain length from C7 to C16 is displayed on the y-axis with the ratio of fatty acid to PHYT (mol:mol), rPHYT, given along the x-axis. Legend: EME (emulsified microemulsion), HII (hexagonal phase), QIID (diamond cubic phase, space group Pn3m), Lc (lamellar crystal).

nanoparticles at 25 °C is presented in Figure 7. The isoprenoid type backbone in phytantriol produces an exaggerated hydrophobe footprint that is not dissimilar to the unsaturated backbone of the monoacylglycerides (“banana” shaped vs “kink”).45 These hydrophobes demonstrate an overwhelming propensity for nonlamellar liquid crystalline phases that are furthermore, prevalent for a broad temperature−composition regime.46,47 Thus, unlike MPO and MO, the addition of Pluronic F127, a steric stabilizer, up to 10 w/w% PHYT does not change the original QIID (Pn3m) phase of the bulk lipid in excess water.12,48 Relative to the monoacylglyceride-FA phase behavior, the incorporation of FA into PHYT yields expanded regions of neat QIID and HII; with the latter becoming dominant as the FA chain length and concentration increased. The EME phase was also present at lower FA chain lengths (C7−C11) and for a marginally broader composition window (rPHYT > 0.43). A relatively narrow QIID + HII coexistence region is present for all FA chain lengths; with Lc + HII region present for C12, C14, C15, and C16 samples broadened with FA chain length. Phase diagrams of the PHYT-FA nanoparticle library at 37 °C, 40 °C, 45 °C, and 50 °C are provided in Supporting Information Figures S10−S13 and S16. With increasing temperature there was a preference for the transition from HII →EME. Specifically at 50 °C, for the PHYT-C7 FA system, the QIID → HII transition occurred at rPHYT = 0.08, with the HII →EME transition at rPHYT = 0.32 (Figure S16); this compares to the system at 25 °C for which rPHYT = 0.48 and 1.27, respectively. The Lc phase was observed to decrease at elevated temperatures due to greater solubilization of long chain FAs in the phytantriol self-assembly objects. In fact, at 50 °C, the only sample in which the Lc phase was detected was PHYT-C16 FA at rPHYT = 1.29. A representative scattering profile of PHYT nanoparticles with the Lc phase is presented in Figure 3B. The LPs of PHYT-FA dispersions are presented in Figure 8. The LP of QIID PHYT dispersions without the addition of FA is around 68 Å, which is smaller than the LP of the corresponding QIID phase in MO (98 Å) and MPO (87 Å). As the FA content in PHYT nanoparticles increased, the QIID lattice reduces to ∼64 Å, at which point a phase transition from QIID → HII occurred. The

LP of the HII phase around the transition is 48 Å, which is also smaller than the LP of HII phases in MO and MPO. 3.4. Coexisting Mesophases and Potential Appearance of Multicompartment Janus Nanoparticles. While the Gibbs phase rule allows for up to four coexisting phases in a quaternary lipid-FA-F127-H2O system, the observation of multicompartment lipid based nanoparticles has only recently been visualized experimentally. To this end, the high throughput screening techniques provide a powerful tool to identify regions of coexisting mesophases, which potentially contain multicompartment nanoparticles. To our knowledge, transient multiphase lipid nanoassemblies have been observed only rarely, though Janus particles with anisotropic surface and bulk properties have been observed in inorganic and polymeric systems. Angelov et al. observed transient multicompartment nanoassemblies containing QIID + QIIP symmetries. It was rationalized that protein−lipid complexation provided the driving force for their organization.22 Individual stable, longlived Janus nanoparticles with biphasic QIID + QIIP and QIID + HII characteristics were recently observed in a MO−capric acid system.23 It was hypothesized that domains with distinct coexisting mesophases were present in the undispersed lipid formulation, with the biphasic character maintained during fragmentation by sonication. In the present study we have set out to further explore the compositional space of related nanoparticle systems containing several different lipids and FAs. Similar to the MO−capric acid phase diagram, the phase behavior of the MO-FA, MPO-FA and PHYT-FA systems exhibit several mixed QIIP + QIID and QIID + HII domains that offer the possibility for Janus nanoassemblies. The coexistence of the two cubic morphologies can be confirmed by comparing the ratio of their lattice parameters as these are theoretically interrelated by the Bonnet transformation, which isometrically maps the underlying minimal surfaces onto each other so that all angles, distances, and areas at all points on the surface are preserved. This predicts that in excess water, and under equilibrium conditions: γP−D = LP(QIIP)/LP(QIID) = 1.28. Table 1 shows the calculated lattice ratios for the phase transitions QIIP → QIID (γP‑D) for MO and MPO (PHYT-FA H

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 8. Lattice parameters of the PHYT-FA nanoparticles at 25 °C as a function of FA to PHYT molar ratio, rPHYT. The FA chain length varies from C7 to C16. The symbols for each mesophase are as follow: QIID (green plus sign), HII (yellow up triangle), EME (red circle), Lc (pink square). Note: for EME, a nonordered L2 phase, the value reported is measured from the single broad peak in the scattering profile, representing the average micelle separation (characteristic distance) of the L2 phase.

exhibited only QIID + HII domains). This corresponds to γP‑D = 1.27- 1.34 for MO, with γP‑D = 1.25- 1.32 for MPO, close to the theoretical Bonnet ratio,35,36 and consistent with previous work that has measured γP−D ∼ 1.33.49 Figure 9 presents a representative Janus lipid nanoparticle of MO and capric acid

(C10) with rMO = 0.06. The LPs measured directly from the cryoTEM images using fast Fourier transform (FFT) were 119 and 94 Å for QIIP and QIID regions, respectively. The lattice ratio is therefore 1.27, close to the value measured by SAXS and the theoretical Bonnet ratio. I

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Table 1. Lattice Ratios at Phase Transition Points from QIIP → QIID (γP−D) for MO and MPO γP−D

MO

MPO

C7 C8 C9 C10 C11 C12 C13 C14 C15 C16

1.33 1.31 1.33 1.34 1.32 1.34 1.27 1.27 1.34 1.32

1.32 1.32 1.32 1.31 1.32 1.32 1.32 1.32 1.25

PHYT-FA systems exhibit several mixed QIIP + QIID and QIID + HII domains, with the Bonnet ratio of the coexisting bcps in the MAG systems γP−D = 1.25−1.34. We believe that these systems offer the tantalizing possibility that populations of compositionally different Janus particles exist. This library of materials will inform QSPR simulations that are being performed in parallel, so that we can predictably formulate and process these types of mixed amphiphile self-assembly materials.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b03769. Partial phase diagrams of MO-FA, MPO-FA, and PHYTFA at temperatures from 37 to 50 °C; temperature dependent partial phase diagrams of MO-FA, MPO-FA, and PHYT-FA nanoparticles; representative scattering profiles of nanoparticle formulation with the Lc phase (PDF)



AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Phone: +61 3 9545 2630. *E-mail: [email protected]. Phone: +61 3 9925 4265. Present Address ∥

N.T.: School of Science, RMIT University, Melbourne, 3000, Australia.

Figure 9. Representative cryo-TEM image of a Janus lipid nanoparticle with coexisting QIIP and QIID phases. The nanoparticle contains MO and decanoic acid (capric acid, C10) with rMO of 0.26.

Notes

The authors declare no competing financial interest.



While we have not explicitly verified the existence of Janus type nanoassemblies using direct visualization methods such as cryoTEM for all samples with different compositions, the similarity of the current compositions with the previously studied MO− capric acid system, together with confirmation of coexisting QIIP + QIID phases from the measured Bonnet ratio, at least presents the tantalising possibility that populations of such biphasic, discrete nanoparticles exist in these formulations. The engineering of such a feature into a well-defined discrete nanoparticle system provides the opportunity to create a wealth of beneficial controllable properties.

ACKNOWLEDGMENTS N.T. is a John Stocker Postdoctoral Fellow sponsored by the Australian Science and Industry Endowment Fund (SIEF). J.Z. is supported by a CSIRO OCE Postdoctoral Fellowship. This research includes work undertaken on the SAXS/WAXS beamline at the Australian Synchrotron, Victoria, Australia. The authors wish to thank Dr. Nigel Kirby and Dr. Stephen Mudie for their support with the SAXS experiments. We thank Dr. Shaun Howard for helping with the operation of the Chemspeed platform and Lynne Waddington for TEM imaging.



4. CONCLUSION In summary, we have established a high-throughput method to produce a large material library demonstrating the influence of saturated FA chain length on the phase behavior and lattice parameter of self-assembled lyotropic lipid nanoparticles produced from a number of cubic phase forming lipids in excess water. The study showed that the carbon chain length and concentration of the FA, and temperature directly affected the mesophase structure of each type of nanoparticle produced. Using this high-throughput method, the influence of other physiochemical parameters, including pH and the charge of biomolecules, to the mesophase structures can be quickly studied. The material libraries produced in this study will also serve as a reference point for the selection of appropriate carriers for drug delivery applications where the phase of the material can play a large role in pharmacokinetic behavior and toxicology.50 Additionally, the phase behavior of the MO-FA, MPO-FA, and

REFERENCES

(1) Boyd, B. J.; Whittaker, D. V.; Khoo, S.-M.; Davey, G. Lyotropic liquid crystalline phases formed from glycerate surfactants as sustained release drug delivery systems. Int. J. Pharm. 2006, 309 (1), 218−226. (2) Chang, C. M.; Bodmeier, R. Swelling of and drug release from monoglyceride-based drug delivery systems. J. Pharm. Sci. 1997, 86 (6), 747−752. (3) Drummond, C. J.; Fong, C. Surfactant self-assembly objects as novel drug delivery vehicles. Curr. Opin. Colloid Interface Sci. 1999, 4 (6), 449−456. (4) Fong, W.-K.; Hanley, T.; Boyd, B. J. Stimuli responsive liquid crystals provide ‘on-demand’drug delivery in vitro and in vivo. J. Controlled Release 2009, 135 (3), 218−226. (5) Mulet, X.; Boyd, B. J.; Drummond, C. J. Advances in drug delivery and medical imaging using colloidal lyotropic liquid crystalline dispersions. J. Colloid Interface Sci. 2013, 393, 1−20. (6) Negrini, R.; Mezzenga, R. Diffusion, molecular separation, and drug delivery from lipid mesophases with tunable water channels. Langmuir 2012, 28 (47), 16455−16462.

J

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(26) Yaghmur, A.; de Campo, L.; Sagalowicz, L.; Leser, M. E.; Glatter, O. Emulsified microemulsions and oil-containing liquid crystalline phases. Langmuir 2005, 21 (2), 569−577. (27) Kirby, N. M.; Mudie, S. T.; Hawley, A. M.; Cookson, D. J.; Mertens, H. D.; Cowieson, N.; Samardzic-Boban, V. A low-backgroundintensity focusing small-angle X-ray scattering undulator beamline. J. Appl. Crystallogr. 2013, 46 (6), 1670−1680. (28) Seddon, J. M.; Squires, A. M.; Conn, C. E.; Ces, O.; Heron, A. J.; Mulet, X.; Shearman, G. C.; Templer, R. H. Pressure-jump X-ray studies of liquid crystal transitions in lipids. Philos. Trans. R. Soc., A 2006, 364 (1847), 2635−55. (29) Briggs, J.; Chung, H.; Caffrey, M. The temperature-composition phase diagram and mesophase structure characterization of the monoolein/water system. J. Phys. II 1996, 6 (5), 723−751. (30) Qiu, H.; Caffrey, M. The phase diagram of the monoolein/water system: metastability and equilibrium aspects. Biomaterials 2000, 21 (3), 223−234. (31) Kulkarni, C. V.; Wachter, W.; Iglesias-Salto, G.; Engelskirchen, S.; Ahualli, S. Monoolein: a magic lipid? Phys. Chem. Chem. Phys. 2011, 13 (8), 3004−3021. (32) Aungst, B. J.; Rogers, N. J.; Shefter, E. Enhancement of naloxone penetration through human skin in vitro using fatty acids, fatty alcohols, surfactants, sulfoxides and amides. Int. J. Pharm. 1986, 33 (1), 225−234. (33) Luzzati, V.; Spegt, P. Polymorphism of lipids. Nature 1967, 215, 701−704. (34) Seddon, J.; Templer, R. Polymorphism of lipid-water systems. Handb. Biol. Phys. 1995, 1, 97−160. (35) Hyde, S. T. Bicontinuous structures in lyotropic liquid crystals and crystalline hyperbolic surfaces. Curr. Opin. Solid State Mater. Sci. 1996, 1 (5), 653−662. (36) Larsson, K.; Tiberg, F. Periodic minimal surface structures in bicontinuous lipid−water phases and nanoparticles. Curr. Opin. Colloid Interface Sci. 2005, 9 (6), 365−369. (37) Israelachvili, J. N. Intermolecular and surface forces., revised third ed.; Academic Press: New York, 2011. (38) Kaasgaard, T.; Drummond, C. J. Ordered 2-D and 3-D nanostructured amphiphile self-assembly materials stable in excess solvent. Phys. Chem. Chem. Phys. 2006, 8 (43), 4957−4975. (39) Cistola, D. P.; Atkinson, D.; Hamilton, J. A.; Small, D. M. Phase behavior and bilayer properties of fatty acids: hydrated 1:1 acid-soaps. Biochemistry 1986, 25 (10), 2804−2812. (40) Cistola, D. P.; Hamilton, J. A.; Jackson, D.; Small, D. M. Ionization and phase behavior of fatty acids in water: application of the Gibbs phase rule. Biochemistry 1988, 27 (6), 1881−1888. (41) Duesing, P.; Templer, R.; Seddon, J. Quantifying packing frustration energy in inverse lyotropic mesophases. Langmuir 1997, 13 (2), 351−359. (42) Shearman, G. C.; Khoo, B. J.; Motherwell, M.-L.; Brakke, K. A.; Ces, O.; Conn, C. E.; Seddon, J. M.; Templer, R. H. Calculations of and evidence for chain packing stress in inverse lyotropic bicontinuous cubic phases. Langmuir 2007, 23 (13), 7276−7285. (43) Templer, R.; Seddon, J.; Duesing, P.; Winter, R.; Erbes, J. Modeling the phase behavior of the inverse hexagonal and inverse bicontinuous cubic phases in 2:1 fatty acid/phosphatidylcholine mixtures. J. Phys. Chem. B 1998, 102 (37), 7262−7271. (44) Rand, R.; Fuller, N.; Gruner, S.; Parsegian, V. Membrane curvature, lipid segregation, and structural transitions for phospholipids under dual-solvent stress. Biochemistry 1990, 29 (1), 76−87. (45) Fong, C.; Le, T.; Drummond, C. J. Lyotropic liquid crystal engineering−ordered nanostructured small molecule amphiphile selfassembly materials by design. Chem. Soc. Rev. 2012, 41 (3), 1297−1322. (46) Hato, M.; Minamikawa, H.; Salkar, R. A.; Matsutani, S. Alkylglycosides with an isoprenoid-type hydrophobic chain can afford greater control of aqueous phase structures at low temperatures. Langmuir 2002, 18 (9), 3425−3429. (47) Hato, M.; Yamashita, J.; Shiono, M. Aqueous phase behavior of lipids with isoprenoid type hydrophobic chains. J. Phys. Chem. B 2009, 113 (30), 10196−10209.

(7) Rizwan, S. B.; Boyd, B. J.; Rades, T.; Hook, S. Bicontinuous cubic liquid crystals as sustained delivery systems for peptides and proteins. Expert Opin. Drug Delivery 2010, 7 (10), 1133−1144. (8) Shah, J. C.; Sadhale, Y.; Chilukuri, D. M. Cubic phase gels as drug delivery systems. Adv. Drug Delivery Rev. 2001, 47 (2), 229−250. (9) Boyd, B. J.; Whittaker, D. V.; Khoo, S.-M.; Davey, G. Hexosomes formed from glycerate surfactantsFormulation as a colloidal carrier for irinotecan. Int. J. Pharm. 2006, 318 (1), 154−162. (10) Zhen, G.; Hinton, T. M.; Muir, B. W.; Shi, S.; Tizard, M.; McLean, K. M.; Hartley, P. G.; Gunatillake, P. Glycerol monooleate-based nanocarriers for siRNA delivery in vitro. Mol. Pharmaceutics 2012, 9 (9), 2450−2457. (11) Barauskas, J.; Johnsson, M.; Joabsson, F.; Tiberg, F. Cubic phase nanoparticles (cubosome): principles for controlling size, structure, and stability. Langmuir 2005, 21 (6), 2569−2577. (12) Dong, Y.-D.; Larson, I.; Hanley, T.; Boyd, B. J. Bulk and dispersed aqueous phase behavior of phytantriol: effect of vitamin E acetate and F127 polymer on liquid crystal nanostructure. Langmuir 2006, 22 (23), 9512−9518. (13) Aota-Nakano, Y.; Li, S. J.; Yamazaki, M. Effects of electrostatic interaction on the phase stability and structures of cubic phases of monoolein/oleic acid mixture membranes. Biochim. Biophys. Acta, Biomembr. 1999, 1461 (1), 96−102. (14) Borné, J.; Nylander, T.; Khan, A. Phase behavior and aggregate formation for the aqueous monoolein system mixed with sodium oleate and oleic acid. Langmuir 2001, 17 (25), 7742−7751. (15) Nakano, M.; Teshigawara, T.; Sugita, A.; Leesajakul, W.; Taniguchi, A.; Kamo, T.; Matsuoka, H.; Handa, T. Dispersions of liquid crystalline phases of the monoolein/oleic acid/Pluronic F127 system. Langmuir 2002, 18 (24), 9283−9288. (16) Yaghmur, A.; Sartori, B.; Rappolt, M. Self-assembled nanostructures of fully hydrated monoelaidin−elaidic acid and monoelaidin−oleic acid systems. Langmuir 2012, 28 (26), 10105−10119. (17) Tran, N.; Mulet, X.; Hawley, A. M.; Hinton, T. M.; Mudie, S. T.; Muir, B. W.; Giakoumatos, E. C.; Waddington, L. J.; Kirby, N. M.; Drummond, C. J. Nanostructure and Cytotoxicity of Self-Assembled Monoolein-Capric Acid Lyotropic Liquid Crystalline Nanoparticles. RSC Adv. 2015, 5 (34), 26785−26795. (18) Yaghmur, A.; de Campo, L.; Salentinig, S.; Sagalowicz, L.; Leser, M. E.; Glatter, O. Oil-Loaded Monolinolein-Based Particles with Confined Inverse Discontinuous Cubic Structure (Fd 3 m). Langmuir 2006, 22 (2), 517−521. (19) Thapa, R.; Baskaran, R.; Madheswaran, T.; Rhyu, J.; Kim, J.; Yong, C.; Yoo, B. Effect of saturated fatty acids on tacrolimus-loaded liquid crystalline nanoparticles. J. Drug Delivery Sci. Technol. 2013, 23 (2), 137−141. (20) Negrini, R.; Mezzenga, R. pH-responsive lyotropic liquid crystals for controlled drug delivery. Langmuir 2011, 27 (9), 5296−5303. (21) Mulet, X.; Gong, X.; Waddington, L. J.; Drummond, C. J. Observing self-assembled lipid nanoparticles building order and complexity through low-energy transformation processes. ACS Nano 2009, 3 (9), 2789−2797. (22) Angelov, B.; Angelova, A.; Filippov, S. K.; Drechsler, M.; Štěpánek, P.; Lesieur, S. Multicompartment Lipid Cubic Nanoparticles with High Protein Upload: Millisecond Dynamics of Formation. ACS Nano 2014, 8, 5216−5226. (23) Tran, N.; Mulet, X.; Hawley, A.; Conn, C. E.; Zhai, J.; Waddington, L. J.; Drummond, C. J. First Direct Observation of Stable Internally Ordered Janus Nanoparticles Created by Lipid Self-Assembly. Nano Lett. 2015, 15 (6), 4229−4233. (24) Mulet, X.; Kennedy, D. F.; Conn, C. E.; Hawley, A.; Drummond, C. J. High throughput preparation and characterisation of amphiphilic nanostructured nanoparticulate drug delivery vehicles. Int. J. Pharm. 2010, 395 (1), 290−297. (25) Mulet, X.; Conn, C. E.; Fong, C.; Kennedy, D. F.; Moghaddam, M. J.; Drummond, C. J. High-throughput development of amphiphile self-assembly materials: fast-tracking synthesis, characterization, formulation, application, and understanding. Acc. Chem. Res. 2013, 46 (7), 1497−1505. K

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir (48) Barauskas, J.; Landh, T. Phase behavior of the phytantriol/water system. Langmuir 2003, 19 (23), 9562−9565. (49) Abraham, T.; Hato, M.; Hirai, M. Polymer-Dispersed Bicontinuous Cubic Glycolipid Nanoparticles. Biotechnology progress 2005, 21 (1), 255−262. (50) Hinton, T. M.; Grusche, F.; Acharya, D.; Shukla, R.; Bansal, V.; Waddington, L. J.; Monaghan, P.; Muir, B. W. Bicontinuous cubic phase nanoparticle lipid chemistry affects toxicity in cultured cells. Toxicol. Res. 2014, 3 (1), 11−22.

L

DOI: 10.1021/acs.langmuir.5b03769 Langmuir XXXX, XXX, XXX−XXX