Physicochemical Characterization of Natural-like Branched-Chain

Article pubs.acs.org/Langmuir

Physicochemical Characterization of Natural-like Branched-Chain Glycosides toward Formation of Hexosomes and Vesicles Noraini Ahmad,†,‡ Roland Ramsch,*,‡ Jordi Esquena,‡ Conxita Solans,‡ Hairul Anuar Tajuddin,† and Rauzah Hashim† †

Department of Chemistry, Faculty of Science, University of Malaya, 50603 Kuala Lumpur, Malaysia Institut de Química Avançada de Catalunya, Consejo Superior de Investigaciones Científicas (IQAC-CSIC), CIBER en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), 08034 Barcelona, Spain

‡

S Supporting Information *

ABSTRACT: Synthetic branched-chain glycolipids have become of great interest in biomimicking research, since they provide a suitable alternative for natural glycolipids, which are difficult to extract from natural resources. Therefore, branched-chain glycolipids obtained by direct syntheses are of utmost interest. In this work, two new branched-chain glycolipids are presented, namely, 2-hexyldecyl β(α)-D-glucoside (2-HDG) and 2-hexyldecyl β(α)-D-maltoside (2-HDM) based on glucose and maltose, respectively. The self-assembly properties of these glycolipids have been studied, observing the phase behavior under thermotropic and lyotropic conditions. Due to their amphiphilic characteristics, 2-HDG and 2-HDM possess rich phase behavior in dry form and in aqueous dispersions. In the thermotropic study, 2-HDG formed a columnar hexagonal liquid crystalline phase, whereas in a binary aqueous system, 2-HDG formed an inverted hexagonal liquid crystalline phase in equilibrium with excess aqueous solution. Furthermore, aqueous dispersions of the hexagonal liquid crystal could be obtained, dispersions known as hexosomes. On the other hand, 2-HDM formed a lamellar liquid crystalline phase (smectic A) in thermotropic conditions, whereas multilamellar vesicles have been observed in equilibrium with aqueous media. Surprisingly, 2-HDM mixed with sodium dodecyl sulfate or aerosol OT induced the formation of more stable unilamellar vesicles. Thus, the branched-chain glycolipids 2-HDG and 2-HDM not only provided alternative nonionic surfactants with rich phase behavior and versatile nanostructures, but also could be used as new drug carrier systems in the future.

■

INTRODUCTION Glycolipids have attracted much attention in the past few decades for their self-assembly properties and as potential biosurfactants.1 These are amphiphilic molecules composed of a sugar moiety and one or several alkyl chains (Figure 1). The amphiphilic character (i.e., hydrophilic and hydrophobic) allows self-aggregation structures at the nanoscale, forming a myriad of liquid crystal phases, from the simple lamellar (Lα), hexagonal (H), or cubic (Q) phases to some less known structures, such as rippled (P) and gel (Lβ) phases. Natural glycolipids are glycoyl derivatives of lipids such as acylglycerols, ceramides, and phenols; collectively these are known as glyconjugates.2 Application of natural glycolipids as biosurfactants has received tremendous attention in recent years with numerous papers and reviews,3−16 due to their unique properties such as mild production conditions, lower toxicity, higher biodegradability, and environmental compatibility. In addition, glycolipid biosurfactants have versatile biochemical properties, are nonionic, and may be derived from renewable resources. Studies on nanoemulsions prepared using natural glycolipids have also been reported extensively.17−22 Thus, © 2011 American Chemical Society

interest in studying these self-assembly systems using novel synthetic glycosides which are closely related structurally to the natural ones is justified. Synthetic glycolipids such as alkyl polyglucosides (APGs) are now industrial materials which have been utilized widely in many applications, from cosmetics and pharmaceutics products to even drilling fluids for oil exploration.23,24 However, these are synthetic monoalkylated glycolipids which will give only the lamellar structure in dry conditions,25 while most natural glycolipids possess double alkyl chains, which are usually asymmetric. It was speculated26 that these features within the branched-chain design promote the formation of liquid crystalline phases with curved interfaces such as the hexagonal and cubic phases, hence supporting cell function. These glycolipids can exhibit thermotropic as well as lyotropic mesophases and therefore they can be considered amphitropic.27 It can be presumed that the synthetic branched-chain Received: September 23, 2011 Revised: December 12, 2011 Published: December 14, 2011 2395

dx.doi.org/10.1021/la203736b | Langmuir 2012, 28, 2395−2403

Langmuir

Article

Figure 1. Examples of natural and synthetic branched-chain glycolipids.

Figure 2. Chemical structures of (A) 2-HDG and (B) 2-HDM. The relative molecular mass of 2-HDG is 404.58 g·mol−1, whereas that of 2-HDM is 566.73 g·mol−1.

glycolipids28 most likely will also give these phases in the thermotropic (dry) structure. Glycolipids influence membrane functionality,29 and therefore, continuous efforts in the synthesis and development of designer synthetic glycolipids are relevant for the improvement of the biomimicking capability of these systems.30 Moreover, synthetic branchedchain glycolipids are suitable, since they represent a simplified model of, and provide a straightforward alternative to, natural glycolipids, which are usually heterogeneous and difficult to extract and to purify.25 Different types of synthetic branched-chain glycolipids (Figure 1) have already been prepared, such as 1,2-dialkyl/ diacylglycerol-based glycolipids,31,32 which have been studied by Mannock et al., and 1,3-glycosylglycerol-based glycolipids, which have been studied by Minamikawa et al.33,34 Another chain design glycoside was introduced recently using Guerbet alcohol by Hashim et al.28 They have produced a large number of compounds from this family by varying the sugar headgroup type (from glucose, galactose, maltose, lactose, etc.) and the chain length (from C8 to C24) using the Lewis acid glycosidation35 in three simple steps. These glycosides were studied for both their thermotropic and their lyotropic phase behavior, aiming to understand the structure relationship to liquid crystal properties.28,26 Some surfactant properties of these materials have been investigated, and these have been shown to be promising for delivery system applications.36,28 In the present investigation, we aim to study the self-assembly and phase behavior of these Guerbet alcohol-based branched-chain glycosides to obtain nanostructured dispersions which could have applications in drug encapsulation and release.

Two synthetic branched-chain Guerbet glycolipids have therefore been selected, which have the same chain length but different headgroups. This allows investigation of the effect of different polarities. They are 2-hexyldecyl β(α)-D-glucoside (2-HDG) and 2-hexyldecyl β(α)-D-maltoside (2-HDM), the chemical stuctures of which are shown in Figure 2. These compounds are both β-dominant (∼90%) anomeric mixtures, which were chosen to avoid the expensive and tedious column chromatography step to isolate the major β-anomer. Although some α/β-glycoside anomers can give different phase behaviors,37 an anomeric mixture (as per our case) was chosen and compared to the behavior of the pure β-glycosides.

■

EXPERIMENTAL SECTION

Materials. The two branched-chain glycosides 2-HDG and 2HDM were synthesized locally28 and further characterized. Anionic surfactants such as dioctyl sodium sulfosuccinate/aerosol OT (AOT) and sodium dodecyl sulfate (SDS) were purchased from Aldrich. Deionized filtered water (Milli-Q, Millipore) with an ionic conductivity of 18.2 μS·cm−1 was used for all sample preparations. Thermal Analysis Determinations. Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) measurements were performed with a Mettler Toledo TGA/SDTA 851e/SF and a Mettler Toledo DSC 821e, respectively. The heating rate was 10.0 °C·min−1 for TGA measurements and 5.0 °C·min−1 for DSC measurements. Data treatment was performed with STARe SW 9.20 software in both cases. Prior to the measurements, the glycolipid samples were dried over phosphorus pentoxide under a high vacuum pump for 24 h to eliminate all moisture. The range for DSC measurement was from −10.0 to +150.0 °C for 2-HDG and from −10.0 to +200.0 °C for 2-HDM. Thermotropic and Lyotropic Study. A Leica Reichert Polyvar 2 optical microscope with a digital Sony CCD-Iris camera was used. 2396


Langmuir

Article

Nonpolarized and polarized lights were applied to study thermotropic and lyotropic behaviors of the novel glycolipids 2-HDG and 2-HDM. A hot stage of the type T95-PE from Linkam was used to control the temperature during the heating and cooling process. Images of the texture were captured and stored using Leica IM500 analysis software. In the thermotropic study, the sample was put on the glass slide and then covered with a coverslip. In the first step, heating was applied to eliminate any moisture trapped in the sample and to form a thin film on the slide. The coverslip was pressed slowly during heating to produce more uniformity on the sample, obtaining better textures. After that, the sample was cooled to 25.0 °C and then heated for the second time. Experimentally, the texture upon cooling was more clearly observed (more defined) compared to that upon heating. The lyotropic experiment (contact penetration), introduced by Lawrence,38 was performed to study qualitatively the formation of liquid crystalline phases in the presence of solvent. A thin film of the surfactant was placed on the glass slide and covered by a coverslip, and a drop of water was placed on the edge of the coverslip. The water penetrates by capillary forces into the sample. The solvated sample was studied under polarized light, and the phases were assessed from the observed textures. Small- and Wide-Angle X-ray Scattering (SWAXS) Determination. X-ray measurements at small and wide angles were performed in a S3MICRO instrument (Hecus X-ray Systems, Graz, Austria) with point focalization, equipped with a GENIX microfocus X-ray source and a FOX 2D point-focusing element (both from Xenocs, Grenoble, France). The scattered intensity (in arbitrary units, AU) was recorded using two position-sensitive detectors (PSDs; Hecus) as a function of the scattering angle defined as θ. The wavelength, λ, was 0.154 nm, and the measurements were performed at 50 kV and 1 mA. 2-HDG and 2-HDM were introduced in glass capillaries (Hilgenberg GmbH) of 80 mm length, 1 mm diameter, and 0.01 mm wall thickness. The SAXS detector covers a range between approximately 0.2° and 8.0°, while the WAXS detector covers a range from approximately 18° to 26°. The temperature controller was a Peltier device. After the capillary was sealed, the usual exposure time for the measurements at 25.0 °C was selected as 30 min, for both dried and hydrated 5.0% 2HDG and 2-HDM. The liquid crystalline phases were characterized by analyzing the relative peak positions when the intensity was plotted as a function of the scattering vector, q, which corresponds to (4π/λ) sin(θ/2). The d spacing of liquid crystalline phases was determined by the Bragg equation (d = 2π/q). Surface Tension Measurement. A tensiometer balance from KRÜ SS with a K12 tensiometer processor has been used for critical aggregation concentration (CAC) determination. The platinum plate method was used for this purpose, with checking of distilled− deionized water prior to measurement. The acceptable required surface tension value for distilled−deionized water is between 71 and 72 mN·m−1. Stock solutions of 2-HDG and 2-HDM were prepared with final concentrations of 0.100 and 0.200 mM, respectively. A series of 2-HDG and 2-HDM solutions with different concentrations were prepared by subsequent dilutions from the stock solution. Binary Phase Behavior Determination. For binary phase diagram determinations, a series of 2-HDG and 2-HDM samples with different concentrations from 0.002 to 0.05 wt % were prepared with a total mass of 2.0 g. All samples were centrifuged for about 5 min at 3000 rpm to ensure that all components were located at the bottom of the tubes. Finally, the tubes were sealed hermetically, were homogenized by stirring, and were placed in a freezer for 24 h. The binary phase determination was conducted in a water bath starting from 5 to 95 °C by monitoring the physical changes as a function of temperature for every 5 °C. Hexosome Preparation. A 0.5 wt % solution of 2-HDG in water was heated at 70 °C for 2 h and then sonicated for 30 min in a PSelecta ultrasound water bath and 15 min with the MS 72 probe of a Bandelin Sonoplus ultrasonic homogenizer (30%, 15 kJ). Vesicle Preparation. A 0.5 wt % solution of 2-HDM in water was heated at 70 °C for 2 h and then sonicated for 30 min in a P-Selecta ultrasound water bath and 15 min with the MS 72 probe of a Bandelin Sonoplus ultrasonic homogenizer (30%, 15 kJ). In the case of

glycolipid mixed with anionic surfactants (SDS and AOT), 5 min of ultrasonication is sufficient for homogenizing the samples. Dynamic and Static Light Scattering. A 3D photon correlation spectrometer (PCS) from LS Instruments was used for dynamic light scattering (DLS) at an angle of 90° and static light scattering (SLS) for angles between 20° and 140°. Multiple scattering is suppressed using 3D cross-correlation technology (down to 5% transmission for a submicrometer sample thickness). The instrument is equipped with a He−Ne laser (632.8 nm). Triplicate readings of 200 s were recorded. The particle (hexosomes and vesicles) radii were calculated by a manual exponential fitting of the first cumulant parameter. For a better visualization, the Contin analysis was performed to give the size distribution, and ALV software from Dullware was utilized. The measurement temperature was maintained at 25 °C by a decalin bath, which matches the refractive index of glass and therefore does not interfere with the measurement. ζ Potential. A Nano-Zetasizer from Malvern was used to study the ζ potential of the glycolipid dispersions. Three measurements of 20 subruns were performed for each sample. ζ potential DTS1060C cells were used to study the samples. A refractive index of 1.40 of liposomes, proposed by the software, was chosen as the reference index. The Smoluchowksi model and automode were applied to treat the data. Cryogenic Transmission Electron Microscopy (Cryo-TEM). Cryo-TEM was performed at the Microscopy Service, Autonomous University of Barcelona, to obtain images of hexosomes and vesicles. The transmission electron microscope was a JEOL JEM 2011 operating at 200 kV. For the preparation of the samples, 5 μL of the sample was deposed on a QUANTIFOIL R 1.2/1.3 grid, and the excess was eliminated with Whatman no. 1 paper. The vitrification was done with a CPC Leica by immersing the grid as fast as possible in liquid ethane.

■

RESULTS AND DISCUSSION Thermogravimetric studies showed that the synthesized compounds of 2-HDG and 2-HDM possessed less than 1 wt % moisture, which was totally removed at 100 °C. 2-HDG started to decompose at 175 °C, whereas 2-HDM decomposed at 225 °C. Differential scanning calorimetry showed that a phase transition of 2-HDG occurred at 57 °C (Figure 3A), and the enthalpy change of first order was calculated as −0.7 ± 0.1 J·g−1 (mean value obtained from three measurements based on the second cycle). On the other hand, 2-HDM exhibits a phase transition at 180 °C, and the enthalpy change was calculated as −1.3 ± 0.1 J·g−1 (Figure 3B). The melting temperature of 2HDM (Mr = 566.73 g·mol−1) was higher than that of 2-HDG (Mr = 404.59 g·mol−1) due to an additional glucose unit in the headgroup of maltoside since the headgroup of the former is formed by combining two glucose units via an α 1 → 4 glycosidic linkage. Increasing the number of sugar units in the headgroup results in an increase not only in its size but also in its hydrogen bonding. Hence, a higher melting point was observed. Furthermore, the transition temperatures for the second heating and cooling cycle also showed slightly smaller values compared to those for the first heating cycle. This can be explained by some minor degradation occurring over the melting temperature. However, the differences are neglectable, and the reproducibility of DSC peaks upon the second heating and cooling indicates that the stability of each compound toward thermal degradation is good. Thermotropic Behavior of 2-HDG and 2-HDM. The behavior of dry surfactant (without solvent) as a function of the temperature can be qualitatively studied by optical polarizing microscopy. Anisotropic phases, such as columnar (hexagonal) or nematic (lamellar) phases, are visible under polarized light. 2397


Langmuir

Article

Figure 5. SAXS spectra of dried (A) 2-HDG and (B) 2-HDM. Figure 3. DSC thermograms of (A) 2-HDG and (B) 2-HDM upon heating and cooling.

of ordering of 2-HDM, and therefore, the structure cannot be considered amorphous. However, the absence of high-order peaks in the SAXS spectra can be attributed to disordered alkyl chains, which produce a relatively less crystalline state, giving rise to the presence of a liquid crystalline phase. It has to be noted that the formation of a smectic A phase has already been observed for a highly pure 2-hexyldecyl β-D-maltoside, examined by synchrotron X-ray diffraction facilities.40 A smectic A phase was formed also by the similar maltosides possessing branched chains.41,34 Thus, the formation of a smectic A phase can be assumed also for the α/β anomeric maltoside. Lyotropic Behavior of 2-HDG and 2-HDM. The investigation of lyotropic liquid crystal phases of the branched-chain alkyl glycosides was carried out using the Lawrence experiment (contact penetration method) with water. In the case of 2-HDG (Figure 6A), two main phases can be

These phases can be identified thanks to their typical patterns. Figure 4 shows two micrographs, under polarized light, of the

Figure 4. Optical polarized micrographs of (A) 2-HDG and (B) 2HDM.

two branched-chain glycolipids. In the case of 2-HDG, a focal conic texture typical of the columnar phase was clearly observed (Figure 4A). Upon heating, the anisotropic phase melted to an isotropic phase between 55 and 60 °C. On the other hand, 2HDM shows a typical texture that corresponds to the smectic A phase (Figure 4B). Upon heating, the anisotropic phase melted to an isotropic phase between 180 and 185 °C. Thus, the melting points of 2-HDG and 2-HDM obtained by optical polarizing microscopy confirmed the melting points determined by DSC. Within error, both results are comparable to those measured previously,36 as well as described for pure βglycoside.28 X-ray scattering confirmed the formation of the columnar phase of 2-HDG (Figure 5A) initially observed under an optical polarizing microscope. The three peaks, indicated by arrows in Figure 5A, possess the typical reflections for a columnar phase, which are 1:√3:√4.39 Using 2d/√3 (for columnar hexagonal), a lattice spacing of 3.0 nm was calculated from the firstorder peak position. Theoretical calculation of the surfactant length of a pure 2-hexyldecyl β-D-glucoside40 gave an overall molecule length of 3.5 nm. In contrast, 2-HDM showed only a single sharp peak in the SAXS spectrum (Figure 5B). The sharp peak indicates a degree

Figure 6. Optical polarized micrographs of the Lawrence experiment for (A) 2-HDG and (B) 2-HDM (L = isotropic phase and HII = inverted hexagonal phase of 2-HDG, whereas L = isotropic phase, Lα = lamellar phase, and Sm A = smectic A phase of 2-HDM). The water gradient decreases from left to right.

observed. From a high to a low water gradient, this corresponds to a phase transition from an isotropic phase (L) to an anisotropic inverted hexagonal phase (HII) at 25 °C. To confirm these results, the system was further investigated by X-ray scattering. A highly concentrated solution of 2-HDG 2398


Langmuir

Article

shown in Figure 7B. In the SAXS spectrum, three equidistant peaks can be observed, which indicates the typical pattern of a lamellar phase. The settled hydrated 2-HDM formed a lamellar liquid crystalline phase with a lattice spacing of 4.3 nm. Table 1 summarizes the results from SAXS experiments for dried and hydrated samples at 25 °C. Critical Aggregation Concentration. The presence of aggregate molecules could be related to the surfactant solutions changing from a clear (one phase) to a turbid (two phases) solution at very low concentrations. As can be seen from Figure

in water was prepared and finally separated by ultracentrifugation. The settled hydrated sample was analyzed by SAXS. Figure 7A shows the SAXS spectrum of the hydrated 2-

Figure 7. SAXS spectra of hydrated (A) 2-HDG and (B) 2-HDM.

HDG at 25 °C. The three peaks show a typical pattern of the hexagonal phase. The lattice spacing is bigger and the patterns are more intense compared to those of the dried sample (Table Table 1. d Spacing and Lattice Spacing (nm) of Dried and Hydrated 2-HDG and 2-HDM Obtained by SAXS Experiments 2-HDG

2-HDM

sample

d spacing (nm)

lattice spacing/repeat distance, a = 2d/√3 (nm)

d spacing (nm)

lattice spacing/ repeat distance, a = d (nm)

dried hydrated

2.5 3.8

3.0 4.4

3.1 4.3

4.3

Figure 8. Surface tension profiles of (A) 2-HDG and (B) 2-HDM solutions as a function of the logarithmic concentration.

8, the surface tensions of these solutions were reduced as a function of the bulk concentration. It is evident that these glycolipid molecules prefer to accumulate at the surface rather than stay solubilized in the bulk solution. This implies that they are highly surface-active materials. The CAC of 2-HDM (0.0085 mM) and that of 2-HDG (0.0070 mM) remain very close (Table 2), although they exhibit the expected tendency.

1), indicating that the hydrated molecules swell further and are much better organized in water.42 This can be due to water molecules strongly bonded to the 2-HDG molecules through H-bonding. Thus, 2-HDG formed a more ordered hexagonal phase with a lattice spacing of 4.4 nm. On the other hand, the optical polarized micrograph of 2HDM (Figure 6B) shows three phase transitions from a high to a low water gradient, namely, isotropic phase (L) → lamellar phase (Lα) → smectic A phase (Sm A), at 25 °C. As can be clearly seen, when water diffused into the solid, the formation of Maltese-cross structures could be observed immediately (highlighted by a white circle). The Maltese-cross structures indicate the presence of a lamellar liquid crystalline phase and moreover vesicle formation is probable. The third phase is the solid sample of the anomeric α/β-maltoside mixture before contact with water and can be assigned to a smectic A phase, as previously observed in the highly pure anomer β-maltoside compound.28 However, X-ray scattering of the settled hydrated 2-HDM at 25 °C confirmed the existence of the lyotropic lamellar phase as

Table 2. CAC Values of 2-HDG and 2-HDM Solutions at 25.0 °C surfactant solution

CAC, mM

ref

2-HDG 2-HDM 2-HDM

0.0070 0.0085 0.0090

this work this work 36

This is due to the difference in their headgroup polarities, although they have the same branched alkyl chain lengths (C16). Indeed, the CAC of 2-HDM is slightly higher because it has a more polar headgroup (bigger headgroup size), giving it a higher ability to aggregate thanks to a larger number of hydrogen bonds. 2399


Langmuir

Article

Binary Phase Behavior of 2-HDG and 2-HDM. The quantitative binary phase diagrams of 2-HDG and 2-HDM in water were studied as a function of temperature. A series of concentrations from 0.002 to 0.050 wt % have been investigated. From the pseudobinary phase diagram of the water/2-HDG system (Figure 9A), a one-phase region

glucose unit in the hydrophilic part does not lead to balanced hydrophilic−lipophilic properties. Thus, an inverted hexagonal phase is formed. Indeed, evaluation of the surfactant packing parameter48 (P) of both surfactants resulted in values of 1.45 and 1.05 for glucoside and maltoside derivatives, respectively (see the Supporting Information). A necessary condition for the formation of reverse structures is a packing parameter P > 1; thus, the value of 1.45 obtained for the glucoside derivative confirms the above assumptions about the formation of an inverted hexagonal phase. Dynamic light scattering measurements of the dispersions (0.5%) showed a hydrodynamic radius of the hexosome of about 100 nm (Figure 10A). Figure 10B shows a micrograph of

Figure 10. (A) Radius distribution of the hexosome dispersion with 0.50 wt % 2-HDG in water obtained by a Contin data analysis of dynamic light scattering results. (B) Corresponding cryo-TEM micrograph of the hexosome dispersion with 0.50 wt % 2-HDG in water. Hexosomes are visible as dark gray, almost spherically shaped particles.

hexosomes measured by cryogenic transmission electron microscopy, which are visualized as hexagonal or spherically shaped particles of about 50−100 nm diameter. As expected, the hexosome particle size measured by cryo-TEM is smaller compared to the particle size measured by photon correlation spectroscopy (Figure 10). This is reasonable, since particle sizes obtained from light scattering are hydrodynamic radii, i.e., the actual particle radius plus strongly bound water, which moves with the object through the bulk phase. As a consequence, the hydrodynamic radius measured by light scattering is usually bigger than that observed by electron microscopy, which shows the actual radius. Moreover, it cannot be excluded that the freezing of the sample, even rapidly performed, shrank the particles, which would also lead to smaller particle sizes. In addition, the relatively high polydispersity index obtained by light scattering measurements (0.3) could indicate the existence of bigger particles or aggregates, which will increase the mean radius of the DLS measurements. The 2-HDG/water dispersions were not stable, and the particle settled down after several days (4−7 days). There are several reasons for the low dispersion stability. First, a weak electrostatic stabilization could be assumed, even though the ζ potential of the hexosome dispersion with 0.50 wt % 2-HDG in water was measured as −33.5 mV, which is usually sufficient for an electrostatic stabilization. Second, the relatively high polydispersity (Figure 10B) significantly influenced the dispersion stability. Finally, the inability of the glycolipid 2HDG to form lamellar bilayers is also a factor in the low dispersion stability. Usually, hexosomes are particles of an inverted hexagonal liquid crystalline structure, stabilized by a layer of surfactant on the water−hexosome interface. Hexosomes based on the hydrophobic surfactant 2-HDG might not be sufficiently stabilized, since the hydrophilic−

Figure 9. Pseudobinary phase diagram of (A) 2-HDG and (B) 2HDM in water as a function of temperature: One-phase region of the isotropic phase (gray) and two-phase region of an inverted hexagonal liquid crystalline phase (A) and lamellar liquid crystalline phase (B) dispersed in water (white).

(isotropic solution, L) was observed from 0.002 to 0.006 wt %, whereas a one-phase region of the water/2-HDM system (Figure 9B) is only observed up to 0.002 wt % at 25 °C. At higher concentration, two-phase regions of both 2-HDG and 2HDM systems have been observed. From optical polarized microscopy and SWAXS results, described above, it can be assumed that, in the two-phase region, 2-HDG could form colloidal dispersions of the inverse hexagonal liquid crystalline phase in the aqueous solution. These dispersions, denominated hexosomes, are known to form in regions of the phase diagram where an inverted type of hexagonal phase (usually denoted HII) coexists in equilibrium with an aqueous solution, made of excess water.42,43 It should be noted that formation of hexosomes has been described in water/glycerol monooleate/ tricaprilin systems,44 although the presence of a stabilizer such as Pluronic 127 was found to improve their stability.42 On the other hand, the two-phase region of 2-HDM can be attributed to a lamellar liquid crystalline (Lα) dispersion in water, which led to the formation of vesicles.45−47 Hexosome Formation by 2-HDG. As mentioned previously, 2-HDG dispersed in water led to the formation of hexosomes, which were composed of an inverted hexagonal liquid crystalline phase. The presence of an inverted hexagonal phase is not yet confirmed, but can be assumed, since the surfactant possesses rather hydrophobic characteristics. The big hydrophobic part with the alkyl branched chain and the single 2400


Langmuir

Article

anionic surfactants (ASs) have been chosen for this purpose, namely, AOT and SDS. For the preliminary test on the influence of anionic surfactants on the aggregation behavior of 2-HDM in water, a ratio of 10:1 (wt %) of 2-HDM to anionic surfactant was chosen. Both 0.50 and 1.0 wt % dispersions of 2-HDM/AOT have been prepared and appeared as more translucent compared to the corresponding 2-HDM dispersion. However, dynamic light scattering measurements (Figure 12A) of the 2-

lipophilic balance of the surfactant does not allow the formation of a stable double layer. Similar cases in the literature described the stabilization of hexosomes with Pluronics, namely, F127.42 Vesicle Formation by 2-HDM. Lamellar liquid crystalline phases form, under high energy input, multilamellar vesicles in water, where the energy input can be in the form of applied shear49 or ultrasonication.45 This work presents lamellar dispersions obtained by the ultrasonication technique. Dynamic light scattering measurements at 25 °C showed a main population at 179 nm in hydrodynamic radius and a wide size distribution. The preliminary result shows that the supplied energy input significantly influenced the size of the vesicles. Higher energy input led to the formation of smaller vesicles, which might be of the unilamellar type. To visualize the vesicles, a lamellar dispersion with 0.50 wt % 2-HDM in water was investigated by cryo-TEM. Figure 11

Figure 12. (A) Radius distribution of a 1.0 wt % 2-HDM/AOT dispersion in water. Two main populations were observed at 80 and 25 nm. (B) Cryo-TEM micrograph of a 1.0 wt % 2-HDM/AOT dispersion. Spherical unilamellar vesicles were observed, indicating that AOT induced the formation of small and large unilamellar vesicles. The bar represents 200 nm.

HDM/AOT mixture show a broad polydispersity, with at least two populations. Indeed, the cryo-TEM micrograph (Figure 12B) shows two types of unilamellar vesicles with different ranges of sizes, which are small unilamellar vesicles (SUVs) with sizes between 20 and 100 nm and large unilamellar vesicles (LUVs) with sizes between 100 and 200 nm. On the other hand, addition of SDS to a 2-HDM dispersion led also to a translucent dispersion with a main hydrodynamic radius of 40 nm, which was measured by dynamic light scattering (Figure 13A). The radius distribution was still broad, Figure 11. Cryo-TEM micrograph of a 0.50 wt % 2-HDM dispersion. MLVs with polydisperse nature can be observed. The bar represents 50 nm.

shows one electron micrograph of multilamellar vesicles (MLVs) in the range of 50−100 nm in radius, which was smaller than the hydrodynamic radius measured by dynamic light scattering. However, the micrographs show high polydispersity. The vesicles are composed of multilayers, which is common for concentrated solutions, such as phospholipids.45,50 The number of layers can be estimated as 3−10, as observed by cryo-TEM, and the interlayer distance can be estimated as around 3.5−4.0 nm, which corresponds to the d spacing measured by SAXS on the hydrated solid surfactant. The stability of the vesicles was very low, and a white precipitate was observed after several days. This can be explained by the high polydispersity, which promotes vesicles fusion, the multilamellar structure, or the temperature dependence of the spontaneous curvature Ho of the surfactant.51 Multilamellar vesicles are obtained by high-energy input, and they are therefore not in thermodynamic equilibrium.45 In addition, low electrostatic repulsion was observed by electrophoretic mobility measurements. The ζ potential of −19.3 mV is too low to stabilize the vesicle formed. To enhance the stability of 2-HDM vesicles, preliminary studies on 2-HDM/ anionic surfactant mixtures were performed. Two standard

Figure 13. (A) Radius distribution of a 1.0 wt % 2-HDM/SDS dispersion in water. The main population was observed at 40 nm. (B) Cryo-TEM micrograph of a 1.0 wt % 2-HDM/SDS dispersion. SDS induced the formation of small unilamellar vesicles. The bar represents 200 nm.

but no other population was detected by DLS. Cryo-TEM micrographs indicated that only SUVs were formed (Figure 13B), with sizes between 30 and 80 nm, which correspond to the hydrodynamic radius measured by dynamic light scattering. The stability of 2-HDM/SDS dispersions was increased for more than one week. The difference between the addition of AOT and SDS might be explained by the different numbers of alkyl chains of the surfactants. While SDS is a monoalkylated surfactant, AOT possesses two alkyl chains in the hydrophobic part. In the aggregation process, the presence of two alkyl chains may disturb more significantly the organization of the 22401


Langmuir

Article

(7) Kitamoto, D.; Ghosh, S.; Ourisson, G.; Nakatani, Y. Chem. Commun. 2000, No. 10, 861−862. (8) Imura, T.; Morita, T.; Fukuoka, T.; Kitamoto, D. Chem.Eur. J. 2006, 231, 2434−2440; presented at the 231st National Meeting of the American Chemical Society, Atlanta, GA, March 26−30, 2006. (9) Garti, N. Colloids Surf., A 1999, 152, 125−146. (10) Franzetti, A.; Gandolfi, I.; Bestetti, G.; Smyth, T. J. P.; Banat, I. M. B. Eur. J. Lipid Sci. Technol. 2010, 112, 617−627. (11) Faivre, V.; Rosilio, V. Expert Opin. Drug Delivery 2010, 7, 1031− 1048. (12) Donnelly, M. J.; Bu’Lock, J. D. J. Chem. Technol. Biotechnol. 1994, 61, 187−195. (13) Brown, M. J. Int. J. Cosmet. Sci. 1991, 13, 61−64. (14) Bednarski, W.; Adamczak, M.; Tomasik, J.; Plaszczyk, M. Bioresour. Technol. 2004, 95, 15−18. (15) Arutchelvi, J.; Bhaduri, S.; Uppara, P.; Doble, M. J. Ind. Microbiol. Biotechnol. 2008, 35, 1559−1570. (16) Albino, J. D.; Nambi, I. M. Funct. Sens. Mater. 2010, 93−94. (17) Yilmaz, E.; Borchert, H.-H. Eur. J. Pharm. Biopharm. 2005, 60, 91−98. (18) Yilmaz, E.; Borchert, H.-H. Int. J. Pharm. 2006, 307, 232−238. (19) Hatziantoniou, S.; Deli, G.; Nikas, Y.; Demetzos, C.; Papaioannou, G. In Microscopy in Nanobiotechnology; Mozafari, M. R., Ed. Micron 2007, 38, 819−823. (20) Desai, A.; Vyas, T.; Amiji, M. J. Pharm. Sci. 2008, 97, 2745− 2756. (21) Deli, G.; Hatziantoniou, S.; Nikas, Y.; Demetzos, C. J. Liposome Res. 2009, 19, 180−188. (22) Clogston, J. D.; Patri, A. K. Methods Mol. Biol. 2011, 697, 109− 117. (23) von Rybinski, W.; Hill, K. Angew. Chem., Int. Ed. 1998, 37, 1328−1345. (24) Balzer, D.; Lüders, H. H. Nonionic Surfactants: Alkyl Polyglucosides; Surfactant Science Series; Marcel Dekker: New York, 2000; p 91. (25) Vill, V.; Hashim, R. Curr. Opin. Colloid Interface Sci. 2002, 7, 395−409. (26) Nguan, T.; Heidelberg, H. S.; Hashim, R.; Tiddy, G. Liq. Cryst. 2010, 37, 1205−1213. (27) Baron, M. Pure Appl. Chem. 2001, 73, 845−895. (28) Hashim, R.; Hassan Abdalla Hashim, H.; Mohd. Rodzi, N. Z.; Duali Hussen, R. S.; Heidelberg, T. Thin Solid Films 2006, 509, 27−35. (29) Goodby, J. W. Liq. Cryst. 1998, 24, 25−38. (30) Cook, A. G.; Wardell, J. L.; Imrie, C. T. Chem. Phys. Lipids 2011, 164, 118−124. (31) Mannock, D. A.; Collins, M. D.; Kreichbaum, M.; Harper, P. E.; Gruner, S.; McElhaney, R. N. Chem. Phys. Lipids 2007, 148, 26−50. (32) Mannock, D. A.; Akiyama, M.; Lewis, R. N.; McElhaney, R. N. Biochim. Biophys. Acta, Biomembr. 2000, 1509, 203−215. (33) Hato, M.; Minamikawa, H.; Seguer, J. B. J. Phys. Chem. B 1998, 102, 11035−11042. (34) Hato, M.; Minamikawa, H.; Tamada, K.; Baba, T.; Tanabe, Y. Adv. Colloid Interface Sci. 1999, 80, 233−270. (35) Vill, V.; Bocker, T.; Thiem, J.; Fischer, F. Liq. Cryst. 1989, 6, 349−356. (36) Hussen, R. S. D. Ph.D. Thesis, Department of Chemistry, Faculty of Science, University of Malaya, Kuala Lumpur, Malaysia, 2010. (37) Auvray, X.; Petipas, C.; Dupuy, C.; Louvet, S.; Anthore, R.; Rico-Lattes, I.; Lattes, A. Eur. Phys. J. E 2001, 4, 489−504. (38) Lawrence, A. S. C. Mol. Cryst. Liq. Cryst. 1969, 7, 1−57. (39) Boyd, B. J.; Whittaker, D. V.; Khoo, S.-M.; Davey, G. Int. J. Pharm. 2006, 318, 154−162. (40) Brooks, N. J.; Hamid, H. A. A.; Hashim, R.; Heidelberg, T.; Seddon, J. M. Liq. Cryst. 2011, 38 (11-12), 1725−1734. (41) Howe, J.; von Minden, M.; Gutsmann, T.; Koch, M. H.; Wulf, M.; Gerber, S.; Milkereit, G.; Vill, V.; Brandenburg, K. Chem. Phys. Lipids 2007, 149, 52−58. (42) Amar-Yuli, I.; Wachtel, E.; Shoshan, E. B.; Danino, D.; Aserin, A.; Garti, N. Langmuir 2007, 23, 3637−3645.

HDM surfactant, which led to the formation of two populations with different radii. On the other hand, the monoalkylated SDS might be more adapted for the issue, allowing the formation of small unilamellar vesicles of a single size distribution. However, further investigation on 2-HDM/AS mixtures will be conducted in the future to optimize the systems' stabilities in view of their application for drug delivery.

■

CONCLUSIONS 2-HDG and 2-HDM are two new natural-like branched-chain glycolipids with interesting/rich phase behavior. The more hydrophobic surfactant, 2-HDG, is characterized by a columnar phase in the dry state, whereas the more hydrophilic surfactant (two glucose units), 2-HDM, shows a smectic A phase. In water−surfactant dispersion, the difference between the two systems is more obvious. 2-HDG forms a hexagonal liquid crystalline dispersion, hexosome, whereas 2-HDM with more balanced hydrophilic−lipophilic properties forms a lamellar liquid crystalline dispersion, which leads to the formation of MLVs. Additions of AOT and SDS to the 2-HDM dispersion induce the formation of unilamellar vesicles, with higher stability. In fact, the two new branched-chain glycolipids possess interesting thermotropic and lyotropic phase properties, as well as binary phase behaviors which lead to the formation of hexosomes and vesicles. This justifies further research on possible incorporation and release of drugs, to be conducted in the future.

■

ASSOCIATED CONTENT

S Supporting Information *

Detailed calculation for the estimation of the packing parameter of both glycosides (2-HDG and 2-HDG). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected].

■

ACKNOWLEDGMENTS Financial support by the InForm Grant-European Union FP7 (Researcher Exchange Program), Consejo Superior de ́ Investigaciones Cientificas (CSIC), Ministerio de Ciencia e Innovación (MICINN; Grant CTQ2008-01979), Generalitat de Catalunya (Grant 2009SGR-961), Postgraduate Research Fund (Grant PS242/2009A), and Grant UM.C/625/1/HIR/ MOHE/05 are gratefully acknowledged. We thank Pablo Castro Hartmann, Technical Services, Universitat Autónoma de Barcelona, for his kind cooperation with cryo-TEM. N.A. extends her sincere thanks to the University of Malaya and Ministry of Higher Education for a Ph.D. fellowship.

■

REFERENCES

(1) Kitamoto, D.; Morita, T.; Fukuoka, T.; Konishi, M.; Imura, T. Curr. Opin. Colloid Interface Sci. 2009, 14, 315−328. (2) Chester, M. A. Pure Appl. Chem. 1997, 69, 2475−2487. (3) Tuleva, B.; Christova, N.; Cohen, R.; Stoev, G.; Stoineva, I. J. Appl. Microbiol. 2008, 104, 1703−1710. (4) Krasowska, A. Postepy Hig. Med. Dosw. 2010, 64, 310−313. (5) Kralova, I.; Sjoblom, J. J. Dispersion Sci. Technol. 2009, 30, 1363− 1383. (6) Kitamoto, D.; Isoda, H.; Nakahara, T. J. Biosci. Bioeng. 2002, 94, 187−201. 2402


Langmuir

Article

(43) Gustafsson, J.; Ljusberg-Wahren, H.; Almgren, M.; Larsson, K. Langmuir 1997, 13, 6964−6971. (44) Amar-Yuli, I.; Garti, N. Colloids Surf., B 2005, 43, 72−82. (45) Gradzielski, M. a. J. Phys.: Condens. Matter 2003, 15, 655−697. (46) Johnson, S.; Bangham, A. Biochim. Biophys. Acta, Biomembr. 1969, 193, 82−91. (47) Johnson, S.; Bangham, A.; Hill, M.; Korn, E. Biochim. Biophys. Acta, Biomembr. 1971, 233, 820−826. (48) Israelachvili, J. N.; Mitchell, D. J.; Ninham, B. W. J. Chem Soc., Faraday Trans. 2 1976, 72, 1525−1530. (49) Diat, O.; Roux, D. J. Phys. II 1993, 3, 9−14. (50) Gentile, L.; Rossi, C. O.; Olsson, U.; Ranieri, G. A. Langmuir 2011, 27, 2088−2092. (51) Bulut, S.; Zackrisson Oskolkova, M.; Schweins, R.; Wennerström, H.; Olsson, U. Langmuir 2010, 26, 5421−5427.

2403


Physicochemical Characterization of Natural-like Branched-Chain

Recommend Documents