Solubilization of a Dendrimer into a Microemulsion - The Journal of

J. Phys. Chem. B 2010, 114, 16723–16730

16723

Solubilization of a Dendrimer into a Microemulsion Ido Nir, Abraham Aserin, Dima Libster,* and Nissim Garti* Casali Institute of Applied Chemistry, The Institute of Chemistry, The Hebrew UniVersity of Jerusalem, Edmond J. Safra Campus, GiVat Ram, Jerusalem 91904, Israel ReceiVed: August 25, 2010; ReVised Manuscript ReceiVed: October 21, 2010

The present work investigates, for the first time, a system comprising a dendrimer incorporated into the water core of water-in-oil (W/O) microemulsion (ME). A second generation (G-2) poly(propyleneimine) dendrimer (PPI) was solubilized into W/O ME composed of AOT (sodium bis(2-ethylhexyl)sulfosuccinate), heptane, and water. Such a model system possessing the benefits of both dendrimers and ME, can potentially offer superior control of drug administration. The localization of PPI within the system, its specific interactions with the components of the carrier, and its effect on the ME structure was explored by SAXS, DSC, ATRFTIR, and electrical conductivity measurements. Considerable water binding by PPI, accompanied by partial dehydration of AOT polar heads, was detected by ATR-FTIR and DSC analysis, suggesting that PPI acted as a “water pump”. In addition, SAXS measurements showed periodicity increase and disordering of the droplets. Hence, localization of PPI within the core and interfacial regions of the droplets was assumed. Direct electrostatic interactions between PPI and the sulfonate group were not noticed, since the dendrimer molecules were mostly not protonated in the current basic environment at pH 12. However, slight hydrogen bonding between PPI and the SdO groups allowed the dendrimer to behave as a “spacer” between sodium and sulfonate ions. This affected the electrical conductivity behavior of the system, revealing that PPI favored the percolation process. Most likely, PPI decreased the rigidity of the interfacial layer, facilitating the diffusion of sodium ions through the channels. The characterized model system can be advantageously utilized to design specific delivery vehicles, allowing administration of dendrimers as a therapeutic agent from host MEs. Introduction Recently, dendrimers have generated considerable interest as potential drug delivery vehicles,1 especially for anticancer therapies2 and diagnostic imaging.3 These globular macromolecules possess very narrow polydispersity and their nanometer size range can allow easier passage across biological barriers. Due to the presence of multiple terminal groups on the exterior of the dendrimer, they provide an excellent platform for the attachment of drugs, cell-specific targeting groups, solubility modifiers, and stealth moieties that reduce immunological interactions.4,5 These macromolecular carriers can also encapsulate hydrophobic drugs inside the nonpolar interior cavities present around the focal core of the dendrimer by a host-guest interaction.6-8 Alternatively, dendrimers by themselves can serve as therapeutic agents by virtue of their activities against prion and Alzheimer’s disease,9 HIV,10 cancer, and others.3 Dendrimers were shown to prevent formation of amyloid fibrils,11 destabilize amyloid aggregates,12 and prevent viral adhesion and replication.10 One of the dendrimers’ major advantages is their ability to pass through the membrane of the target cell with the encapsulated drug. However, dendrimers suffer from relatively low encapsulation capacity of various drugs and dermal penetration enhancers, compared to colloidal carriers such as microemulsions (MEs). Moreover, it seems that the potential capability to sustain delivery of the encapsulated drugs is much lower, compared to colloidal vehicles. To improve the performance * Corresponding authors. N.G.: tel, +972-2-658-6574/5; fax, +972-2652-0262; e-mail: [email protected]. D.L.: tel, +972-2-658-4962; fax, +972-2-652-0262; e-mail: [email protected].

of dendrimers as delivery vehicles, we propose an alternative type of drugs administration, based on immobilization of dendrimers into an ME. Microemulsions are optically isotropic and thermodynamically stable nanostructured mixtures of water, oil, and amphiphile(s), typically sized in the range 10-100 nm.13,14 These systems frequently require cosolvents or cosurfactants to achieve very low interfacial tension and to facilitate proper packing parameters. The most common ME structures are classified as water-in-oil (W/O), bicontinuous, and oil-in-water (O/W).15 Due to their special features, MEs offer several advantages for drug delivery, including ease of preparation, thermodynamic stability, and improved drug solubility and bioavailability. Consequently, MEs have been extensively studied as drug delivery vehicles.16 In this paper, for the first time, we propose a complex drug delivery system, based on a dendrimer solubilized in the aqueous core of a W/O ME. This approach may combine the advantages of both dendrimers and MEs, to provide better control of drug release. To demonstrate the feasibility of our concept, we utilized a W/O ME, composed of AOT (sodium bis(2-ethylhexyl)sulfosuccinate), heptane, and water, as a model and a wellcharacterized host system for solubilization [second generation (G-2) poly(propylene imine) dendrimer (PPI)]. AOT (Figure 1a) is a frequently used anionic surfactant that forms stable W/O MEs with a series of oils without the need for a cosurfactant. It has been demonstrated that AOT molecules assemble into welldefined spherical aggregates with the droplet size changing proportionally with water content.17 PPI polycationic dendrimers are of particular interest since they are biocompatible and commercially available.1,7 Another probable benefit of such dendrimers is the fact that polycationic dendrimers were shown

10.1021/jp108040y  2010 American Chemical Society Published on Web 12/02/2010

16724

J. Phys. Chem. B, Vol. 114, No. 50, 2010

Nir et al.

to react with bacterial membranes and disturb their integrity.18 The PPI dendrimer of second generation (Figure 1b) possesses eight primary amine groups on the surface and has a spherical shape with a radius of gyration of 6 Å.19 In aqueous solution at pH < 9.85, most of the PPI G-2 terminal amine groups are protonated and have positive surface charges.20,21 The present study aims to explore a model drug delivery system composed of a PPI dendrimer entrapped in an AOTbased W/O microemulsion. The effect of the dendrimer solubilization on the ME structure, and the host-guest interactions, were elaborated using small-angle X-ray scattering (SAXS), differential scanning calorimetry (DSC), attenuated total reflectance Fourier transform infrared (ATR-FTIR), and electrical conductivity measurements.

distance was calibrated using silver behenate and was 1840.5 mm. X-ray data analysis was performed as described by Ezrahi et al.15 Differential Scanning Calorimetry (DSC). A Mettler Toledo DSC822 (Greifensee, Switzerland) system was used for calorimetric measurements. The instrument was calibrated with indium, lauric acid, water, and ethyl acetate to ensure the accuracy of the caloric data. The heating rate of calibration was 10 K min-1. The DSC measurements were carried out as follows: 7-11 mg microemulsion samples were weighed, using a Mettler M3 microbalance, in standard 40 µL aluminum pans and immediately sealed by a press. The samples were cooled in liquid nitrogen from +25 to -100 °C at 6 °C/min. Each sample remained at this temperature for 20 min and was then heated at a rate of 5 °C/min to 25 °C. An unloaded pan was used as a reference. The enthalpy changes associated with the thermal transition were obtained by integrating the area of each pertinent DSC peak. DSC temperatures reported here were reproducible within (0.5 °C. We followed the method used by Senatra et al.22 to identify various states of water in our systems, as described elsewhere.15,23 Attenuated Total Reflectance Fourier Transform Infrared (ATR FTIR) Measurements. An Alpha T model spectrometer, equipped with a single reflection diamond ATR sampling module, manufactured by Bruker Optik GmbH (Ettlingen, Germany), was used to record the FT-IR spectra. The spectra were recorded with 50 scans, at 25 °C; a spectral resolution of 2 cm-1 was obtained. Multi-Gaussian fitting has been utilized to resolve individual bands in the spectra. The peaks were analyzed in terms of peak frequencies, width at half-height, and area. Electrical ConductiWity (EC). EC measurements were performed at (1-60) ( 0.2 °C using a conductivity/temperature meter (Extech Instruments, Waltham, MA, U.S.). A continuous stirring was carried out during the measurements. The temperature and the conductivity were simultaneously detected.15

Experimental Methods

Results and Discussion

Materials. Sodium bis(2-ethylhexyl)sulfosuccinate (99% purity) and n-heptane (>99% purity) were purchased from Sigma Chemical Co. (St. Louis, MO). PPI (second generation) (>95% purity) was obtained from SyMO-Chem Netherlands. Methods. Preparation of the Microemulsions. The AOTbased microemulsions were prepared by adding the aqueous phase (solution of PPI-G2 in distilled water) to a binary mixture of AOT and heptane (1:1.5 weight ratio, respectively). The aqueous phase content in the microemulsion samples was always 24 wt %, while the PPI content in the aqueous phase was 0-25%. Small-Angle X-ray Scattering (SAXS). SAXS measurements were used to identify the structure of the microemulsions containing various quantities of dendrimer (0-25% in the aqueous phase). The scattering experiments were performed using Cu KR radiation (λ ) 0.154 nm) from a Rigaku RAMicroMax 007 HF X-ray generator operated at a power rating up to 1.2 kW generating a 70 × 70 mm2 spot size and focus. The direct beam then goes through a vacuum Osmic CMF12100CU8 focus unit and, defined by a set of two scatterless slits, the beam size at the sample position is 0.7 × 0.7 mm2. The scattered beam goes through a flight path filled with He and reaches a Mar345 Image Plate detector. The sample was inserted into 1.5 mm quartz capillaries that were then sealed. The measurements were performed at room temperature (25 ( 2 °C) with an exposure time of 0.5 h. The sample to detector

PPI was solubilized into the ternary of the AOT/heptane/water system, at constant aqueous phase content of 24 wt %, while the dendrimer concentration was varied from 5 to 25 wt % of the aqueous phase. At higher PPI content the MEs were unstable, separating to a two-phase solution. It should be noted that solubilization of PPI into the ME at pH < 9.85 was impossible, since the addition of the dendrimer hampered formation of stable MEs in such conditions. The probable reason is the formation of a strong complex between the positively charged dendrimer and negatively charged sulfonate. Hence, we worked at conditions (pH ) 12) providing unprotonated terminal amine groups of PPI, modeling a situation where the dendrimer is a therapeutic agent. This was done to reduce potential ionic interactions of PPI with the surfactant within the micelles. Once delivered into physiological conditions (pH 7.4), such a dendrimer should regain a positive surface charge, which is beneficial to the membrane permeation. As shown in Figure 1b, PPI molecules involve two sorts of amine groups: the primary amine end groups on the periphery of the dendrimer and the tertiary amine groups located at the branching points and the core of the molecules. These two types of amine groups, if isolated, are characterized by different pK values, which are the negative logarithms of the acidic dissociation constant for the protonated primary (pKp) and tertiary (pKt) amine groups, correspondingly. The basicity of the primary amine group is higher compared with that of the tertiary one, i.e., pKp > pKt. For PPI-G2,

Figure 1. Schematic chemical structure of (a) the surfactant (AOT) and (b) the dendrimer (PPI G-2).

Solubilization of a Dendrimer into a Microemulsion

J. Phys. Chem. B, Vol. 114, No. 50, 2010 16725

Figure 2. Typical experimental SAXS curves of W/O blank and PPI loaded microemulsions, measured in blank ME (O) and in the presence of 20 wt % PPI (∆).

TABLE 1: Periodicity (d) and Correlation Length (ξ) Calculated from the SAXS Measurements upon PPI Solubilization in the ME System PPI concn (wt %)

d (Å)

ξ (Å)

0 10 20 25

85 88 94 98

393 356 325 307

Kabanov et al.20 estimated (pKp) and (pKt) as 9.85 and 6.0, respectively. Thus, the degree of protonation, even of the primary external amine groups is negligible (∼1%), meaning that most of the PPI molecules in the W/O droplets are not charged at all. To investigate the phenomenological effect of PPI solubilization on the ME structural dimensions and water behavior in the droplets, SAXS and DSC measurements were applied. SAXS Measurements. Typical SAXS diffractions of the empty MEs and a PPI loaded sample (at 20 wt %) at 25 °C are displayed in Figure 2. As shown, broad diffraction peaks were obtained, which are typical for the micellar system. In the presence of PPI the peak broadened and shifted toward lower q (nm-1) values (Figure 2). Table 1 shows the dependence of the periodicity, d, and correlation length (persistence length, reflecting the degree of order in the microemulsion), ξ, as a function of PPI content (0-25 wt % of the aqueous phase), which were calculated after fitting the scattering patterns of the samples to the Teubner and Strey equation.24 Moderate swelling of the ME droplets and a decrease of the order degree of the droplets were monitored. It was found that at the largest PPI concentration (25 wt %) the periodicity value increased by 15%, while the correlation length decreased by 20%, compared to that for the empty ME. This trend may be attributed to at least partial intercalation of the dendrimer into the interfacial region, thereby inducing the slight increase of the periodicity and disordering of the reversed micelles. DSC Measurements. To follow the state of the water of the reversed micelles (RM) upon PPI solubilization, DSC was implemented. This technique is widely used for studying the low-temperature behavior of RM systems, including AOT-based MEs.25 To evaluate the degree of binding strength that varied with PPI solubilization, the equations used by Senatra et al.22 and Ezrahi et al.15,26 to identify various states of water were applied. The thermal behavior of the studied systems in low and high PPI content is represented in Figure 3a,b, respectively. In the dendrimer-free system the endothermic peak (-0.45 °C) was assigned to the free water fusion. Generally, endothermic events attributed to water behavior can be related to the formation of three types of water: bound water (which is associated with hydrophilic groups and melts below -10 °C),

Figure 3. DSC thermograms of the W/O MEs with (a) low PPI content (0-10 wt % from the aqueous phase) and (b) high PPI content (15-25 wt % from the aqueous phase).

interphasal water (defined as water confined within the interface of the dispersed system, which melts at about -10 °C), or free water (which melts at about 0 °C).15,26 To verify that the discussed endothermic events are related to water, we replaced the water with D2O, leaving all other components unchanged, which resulted in a characteristic shift to higher temperatures (data not shown). The fusion temperatures of water molecules are depicted in Figure 3a,b and summarized in Table 2. It is evident that the embedment of the dendrimer resulted in a gradual decrease of the fusion temperatures from -0.45 °C in the empty systems to the lowest temperature of -11.7 °C at 25 wt % PPI (Table 2). The drop in Tfusion was also accompanied with pronounced peak broadening and low melting enthalpies, corresponding to stronger binding of water molecules. This indicates that with increasing PPI concentration, more water molecules become bound to its peripheral amine groups, transforming free water to interfacial. It should be noted that ∼63 wt % of the water in the empty ME was classified as free water, available for PPI solubilization. While the solubilization of 5 wt % PPI caused no detectable changes in the water behavior, higher dendrimer content (10-20 wt %) induced formation of broad peaks containing interfacial water, coexisting with free water. This trend was mostly evident in the samples with 10 and 20 wt % of PPI where the water fusion peaks consisted of two subpeaks, reflecting various populations of water. Although it was practically impossible to separate and quantify these different proportions of water populations, such a tendency suggested the existence of different water types, varying by binding strength to the dendrimer. Finally, at PPI concentration of 25 wt % no free water was detected, but only interfacial (12.5 wt %) and nonfreezable water (87.5 wt %). Substantial water binding revealed by DSC measurements throughout the investigated interval of PPI concentrations

16726


Nir et al.

TABLE 2: Thermal Behavior of the ME System upon Varied PPI Content (0-25 wt %) of the Aqueous Phasea PPI concn (wt %)

Tfusion (°C)

∆H (J/g)

0 5 10

-0.4 -0.5 -3.7 -1.4 -2.4 -7.6 -3.2 -11.7

205 197 124

15 20 25

free water (%) 62.6 60.2

110 85 39

interfacial water (%)

nonfreezable water (%)

not detected not detected free + interfacial water

37.4 39.8 b

free + interfacial water free + interfacial water

b b

not detected

12.5

87.5

a

Melting temperatures and the corresponding peak enthalpies are tabulated. Variation in the content of free, interfacial, and nonfreezable water are shown. b Could not be calculated.

Figure 4. Gaussian components of the O-H stretching vibration of H2O in the empty W/O. The components are located at surfactant bound, 3587 cm-1, high angle, 3462 cm-1, and low angle, 3318 cm-1.

indicated that PPI functioned as a “water pump”. This observation is also consistent with SAXS analysis, both suggesting intercalation of PPI within the core and interfacial regions of the droplets, which induced some degree of periodicity (d) increase and disordering of the droplets. To get more detailed molecular understanding of the PPI interactions with the ME components, ATR-FTIR analysis was applied. A number of ATR-FTIR studies conducted on W/O AOT-based MEs demonstrated the marked capability of this methodology for detailed analysis of specific molecular interactions within RM.27-29 ATR-FTIR Analysis. In the present study, we analyzed the investigated system in four distinct structural regions, characterizing water (OH stretching bands) and AOT (symmetric and antisymmetric sulfonate stretching modes and the carbonyl stretching region). Water Stretching Modes. The vibrational stretching modes of water are very sensitive to the environment and intermolecular interactions.30,31 In this particular case, the stretching modes of water were used to understand the type of bonding between water molecules and the PPI dendrimers. The water state of reverse micelles usually consists of a two-population system, including a bound water layer associated with the surfactant surface and a core of bulk-like water. It should be noted that FTIR information is not equivalent to the thermal analysis and may be compared with DSC data only qualitatively. It was demonstrated that fitting the O-H stretching vibration to a series of three Gaussian functions enables quantitative insight of the different water types distribution within the reverse micelle of AOT microemulsions.29,30 In accordance with this procedure, the O-H stretching vibration of the empty microemulsion was fitted to three Gaussian components centered at 3587, 3462, and 3318 cm-1 (Figure 4). Consistent with the procedure developed by Brubach et al.,31 the high frequency component was assigned to the bound water, which is made up of water molecules either solvating the surfactant headgroups (highly strained H-bonds) or trapped in the micellar interface (few or no H-bonds). The two lower frequency components are attributed to bulk-like water with moderately strained H-bond

Figure 5. O-H stretching vibrations of H2O in the W/O microemulsion as a function of PPI content (wt %) from the aqueous phase.

angles (intermediate frequency) and unstrained H-bond angles (low frequency).11 Incorporation of PPI into the ME induced significant shift of all three υ(OH) positions toward lower wavenumbers (Figure 5), shifting from 3587 to 3577 cm-1 for the bound water (Figure 6a), from 3462 to 3455 cm-1 for bulklike water with moderately strained H-bond angles (Figure 6b), and from 3318 to 3290 cm-1 for bulk-like water with unstrained H-bond angles (Figure 6c). These shifts provided direct evidence that both bulk-like water and bound water populations were strongly affected by the intercalation of PPI to the water pool of the reversed micelles. It was reported that in the AOT-based reversed micelles the influences of headgroups and counterions on water structure are opposite.29 The hydration of anionic surfactant headgroups tends to increase the electron cloud density of hydrogen atoms in water molecules, with consequent breaking of hydrogen bonds of bulk-like water. The strength of the O-H bond increases accordingly. This factor makes the water O-H stretching vibration be located at higher frequencies. On the contrary, counterions, accumulating in the cores of reverse micelles, which can polarize water molecules, lead to a lower electron cloud density around the protons and a decrease in intensity of the O-H bond of water. This effect causes the O-H stretching band to center at lower frequencies. Following this interpretation, it is reasonable to suggest that the dendrimer decreased electron cloud density around the bulk-like water protons, thereby weakening the O-H bond of water, inducing a transition to lower wavenumbers. The same phenomenon of weakening the O-H bonds was also noticed in the bound water region. From that, one can conclude that competition for water


Figure 6. Frequency ranges (cm-1) of Gaussian components fit to the O-H stretching modes of reverse micellar water pools at varied PPI concentrations for the aqueous phase: (a) surfactant bound component; (b) high angle component; (c) low angle component.

binding between PPI and AOT sulfonates, which resulted in partial dehydration of the latter, is responsible for the transition of the bound water IR component (3587 cm-1 band) to lower frequencies. It is also well documented32 that in addition to sulfonates, water also interacts with the AOT polar head through H-bonds with CdO groups. Hence, the dehydration process of AOT polar moieties may be reflected as well on the carbonyls, as discussed below. Carbonyl (CdO) Stretching Mode. The carbonyl bands of AOT upon solubilization of PPI are depicted in Figure 7a. Using a two-Gaussian decomposition of the CdO stretching band data, it was observed that the carbonyl stretching mode exhibited two clearly separated peak maxima (Figure 7b), at 1739 and 1726 cm-1, indicating that these groups are related to two different conformations about the acyl CsC bond of the succinate backbone. According to Moran et al.,27 the high frequency band was assigned to the approximately gauche conformation and the lower one (at 1726 cm-1) was related to the approximately


Figure 7. (a) CdO stretching vibrations of AOT in the W/O microemulsion as a function of PPI content from the aqueous phase. (b) Gaussian components of the carbonyl stretching vibration of H2O in the empty W/O microemulsion. The 1735 cm-1 band was assigned to the approximately gauche conformation and the lower one (at 1726 cm-1) was related to the trans conformation. (c) Normalized peak areas A1726/(A1726 + A1739) as a function of PPI concentration from the aqueous phase.

trans conformation. While the wavenumber positions of the carbonyl stretching mode remained intact, a gradual decrease in the intensity of the 1726 cm-1 band relative to the 1739 cm-I band was detected (Figure 7a). To account for concentration changes as the PPI content was increased, the normalized peak areas A1726/(A1726 + A1739) were calculated (Figure 7c). Here it is demonstrated that the relative populations of trans isomers decreased in favor of the increase of the gauche isomers content, reflecting a change in the relative energies of the different conformations about the acyl CsC bond. This was attributed to a decrease in the average area per headgroup observed upon dehydration of carbonyls. Such conformational modifications, reflected by the changes of normalized areas of CdO low and high frequency components, were detected in the hydration and

16728


Nir et al.

Figure 8. Symmetric stretching bands of AOT in the W/O microemulsions at varied PPI concentration from the aqueous phase.

dehydration processes of the carbonyl groups of ester linkages of AOT, with water molecules.27 Again, the most reasonable explanation for the partial dehydration of the AOT carbonyls is a competition for the water binding between the dendrimer and the polar headgroups of the surfactant, which was also detected in the stretching modes of water. Moreover, it is important to note that a drastic increase of the observed dehydration of CdO groups (Figure 7c) occurred in ME samples with PPI content greater than 12.5 wt %. This finding is in agreement with the downward shift of highest energy Gaussian contribution of the water stretching band (bound water) (Figure 6a), where the same onset of the critical behavior (12.5 wt % PPI) was noticed. A stronger decrease of this OH stretching band maximum was observed at greater content of PPI (12.5-25 wt %), compared to the lower dendrimer concentrations (up to 12.5 wt %). It follows from these results that the dendrimer incorporation into the amphiphile-water interfacial region turned out to be more significant only at high solubilization loads (above 12.5 wt %). Probably, below this solubilization content the dendrimer was mostly intercalated in the bulk water region, where the water activity was high. PPI intercalation into the interface presumed competition for interfacial water binding with the polar headgroups of AOT. Such an effect was indeed shown by both the perturbation of the bound water by PPI and the dehydration of CdO groups of the surfactant. Furthermore, to determine whether direct interaction of PPI with AOT sulfonate groups took place, SdO symmetric and antisymmetric stretching modes were analyzed. Symmetric SdO Mode. The SdO stretching bands were reported to be sensitive to electrostatic interactions between the Na+ counterion and the SO3- group.32,33 Hence, this region provides additional information on possible AOT polar head interactions with the dendrimer. The symmetric SdO stretching bands are presented in Figure 8. The symmetric SdO maximum was found to shift up to 2 cm-l (from 1046 to 1044 cm-l) in the presence of PPI. It was claimed in the literature that the shift of the υ(SO3-) mode depends strongly on the water content in the droplet, since electrostatic cation-anion interactions are affected by charge-dipole interactions between water and the sodium counterion.32,33 Besides, the position of this peak is highly sensitive to the alkali-metal cation.34 Taking this into consideration, the small change observed was assigned to a slight weakening of the sodium (cation)-sulfonate (anion) interaction upon PPI incorporation, which may result in a partial exclusion

Figure 9. (a) Infrared spectra of AOT antisymmetric sulfonate in the range 1130-1330 cm-1 at selected values of PPI. (b) Curve-fitted results of infrared spectra of 1130-1330 cm-1 of the empty microemulsion.

of the sodium counterion from the sulfonate headgroup. Minor interaction between slightly protonated PPI (∼1% of amines are charged) and anionic sulfonate may take place, thus somewhat decreasing the interactions between sodium and sulfonate. Another possible reason for such weak interactions is the hydrogen bonding between the external amine groups of PPI and the sulfonate groups, which are responsible for a partial removal of the sodium counterion from the SdO headgroup. In this case noncharged PPI might slightly separate between sodium and sulfonate ions. Antisymmetric SdO Mode. The antisymmetric vibration mode of AOT in the frequency range 1130-1320 cm-1 was also considered (Figure 9a). This broad profile is composed of the absorbance of the asymmetric SdO stretching mode, the CsO and CsC stretching band of the ester bond, as well as the CH2 twisting mode.27,28 Of our special interest in this range are the antisymmetric sulfonate stretching modes that evolved as a doublet at ca. 1209 and 1242 cm-l and the strong mode at approximately 1160 cm-1, assigned to a combination of CsO and CsCC stretching modes of the ester linkage (Figure 9b). The doublet indicates that the degeneracy of this vibration is lifted by a nonsymmetric interaction of the sodium cation with the SO3- headgroup. In other words, the magnitude of the difference in wavenumber of the two split peaks reflects the interaction between the sodium cation and sulfonate group. It was demonstrated that the magnitude of the band splitting decreases with increasing micelle hydration due to a weakening of the sodium-sulfonate interaction as a result of H-bond interaction between water and the AOT polar headgroup with the corresponding increases in sodium-sulfonate spatial separation.28 In our experiments no significant change was detected in the splitting of the doublet, keeping the magnitude of the split around 31.8-33 cm-1. This result suggests that no significant interaction between PPI and the AOT sulfonate was noticed. Another plausible reason for this observation is that



Figure 10. Electrical conductivity as a function of temperature for reverse micellar systems at different PPI concentrations from the aqueous phase: (0) blank; (∆) 5 wt %; (O) 10 wt %; (+) 20 wt %.

there is a large excess of AOT over PPI molecules. Therefore, apparently most of the surfactant molecules did not interact with the dendrimer. Furthermore, this conclusion was supported by a gradual but only slight shift (up to 1.5 cm-1) of CsO and CsCC stretching modes of the ester linkage from 1164.6 cm-1 in the empty ME to 1163 cm-1 at maximum PPI loading. This means that just a weak direct interaction of PPI with the AOT ester moiety was detected. Electrical Conductivity Measurements. The effect of the dendrimer presence on the ME electrical conductivity and in particular on its percolation behavior is of great interest, since it may clarify the nature of the interaction between PPI and the interface of the droplets. Moreover, since percolation has a significant impact on transport processes within the MEs,17 it may also affect diffusion of the encapsulated dendrimer. The driving force of the percolation in an ME is the interaction between water droplets. During the percolation process water droplets coalesce into clusters, and ions either hop from droplet to droplet or are transferred by way of transient diffusion. As a consequence of the ion transfer, the conductance (σ) can be enhanced 100- to 1000-fold. The point of maximum gradient of the profile corresponds to the transition of the percolation process and is designated as the threshold temperature (Tp). Figure 10 presents the effect of PPI incorporation on the temperature-dependent conductivity of the MEs. Variation of conductivity with temperature for the empty ME showed that, at low temperature, the conductivity remained constant. Once approaching the percolation transition, the conductivity increased markedly, showing a sigmoidal behavior. Tp established at 43 °C for the empty ME, sharply decreased to 22 °C in the sample loaded with 5% PPI and to 15 °C in the presence of 10 wt % dendrimer. For 20 wt % PPI the Tp should be approximately positioned at 3 °C. Hence, in the studied concentration range, PPI favored percolation. As has been established,35 percolation is hindered by additives stiffening the surfactant membrane and favored by additives that make the membrane more flexible. It was reported that a group of molecules such as urea,36 formamide,37 ethylene glycol, and polyethylene glycols (PEGs)38 favored the electrical percolation phenomenon in MEs. These additives acted as “spacers” within the surfactant monolayer, thereby decreasing its spontaneous curvature and facilitating droplets clusterization and the formation of ion transport channels between droplets upon collision. At high additive concentration replacement of water molecules from the interface by those additives occurred, further facilitating the formation of ion transport channels between droplets. Similar to the mentioned type of additives, the PPI dendrimer affected the

Figure 11. Schematic presentation of the PPI dendrimers encapsulated within W/O microemulsions: within a single droplet and within a three droplet cluster. The dendrimer interaction with the AOT polar heads locally decreases the interfacial rigidity and thereby enhances droplets attraction and the formation of channels between the droplets.

interfacial layer of the MEs, consequently favoring the percolation process. This finding is in line with our previous analysis done by FTIR and SAXS, demonstrating that PPI, due to its inclusion in the AOT interface, replaced water molecules from the hydration sphere of the AOT headgroup. The observed increase in the droplet size upon addition of PPI (Table 1) indicated that the surface area per AOT molecule increased. Considering the detected dehydration of the surfactant, the only reason for this phenomenon is the incorporation of PPI molecules as “spacers” between AOT molecules. Such a process, combined with dehydration of AOT, increased interfacial disorder and deformability, and probably decreased the rigidity of the interfacial layer (Figure 11). Hence, both the dehydration of AOT headgroups and intercalation of PPI into the interface region as a spacer, decreased the rigidity of the interfacial film; favored the transport of the Na+ through the channels, enhancing easier fusion; and lowered the percolation temperature. Conclusions In the current study, we investigated for the first time a system comprising dendrimer encapsulated in the water core of W/O microemulsion. It was demonstrated that at constant aqueous phase content (24 wt %) up to 25 wt % of PPI (from the aqueous phase) could be solubilized into the ionic AOT-based system. The impact of PPI dendrimer incorporation into the reversed micelles was characterized by SAXS, DSC, ATR-FTIR, and electrical conductivity measurements. Followed by PPI solubilization, structural changes in the W/O droplets were reflected by the increase in the periodicity values (from 85 to 98 Å) and the decrease of the correlation length (from 393 to 307 Å), at maximum PPI loading.

16730


Monitoring the thermal behavior of water in these systems revealed that the water fusion endothermic event moved progressively to lower temperatures as the PPI concentration increased. Water was gradually transformed from free to interfacial, in particular at high PPI content (15-25 wt %). This was also confirmed by broadening of the peaks, appearance of subpopulations of interfacial water, and a decrease of the melting enthalpies. Water binding to amine groups of PPI was suggested and further confirmed by ATR-FTIR analysis of υ(OH) water stretching bands. Considerable transitions to lower wavenumbers of the bound and bulk-like water components, found from a Gaussian decomposition, led us to conclude that PPI binds water both within the core and at the interfacial region of the swollen micelles. A competition for water binding between PPI and AOT polar moieties was reflected as well on the carbonyls of the surfactant. The CdO stretching mode was also found to bear significant modifications upon PPI inclusion. The relative population decrease of trans isomers in favor of an increase of the gauche isomer content was assigned to a decrease in the average area per headgroup observed upon dehydration of carbonyls. Onset of the critical behavior (12.5 wt % PPI) was monitored and reflected a strong increase of the observed dehydration of CdO groups and more pronounced water binding in the interfacial region. It was suggested that the PPI presence in the AOT-water interfacial region was more significant at a high concentration range, 12.5-25 wt %. This imposed competition for interfacial water binding between PPI and AOT. Examination of symmetric and asymmetric SdO stretching bands of AOT indicated that there was no direct measurable electrostatic interactions between PPI and the sulfonate group. Weak hydrogen bonding between PPI and the SdO groups was suggested, enabling PPI to function as a “spacer” between sodium and sulfonate ions. Such a situation was imposed by the fact that most of the PPI molecules were not protonated in the present conditions at a pH of 12. The observed modifications in the interface of the droplets radically influenced the electrical conductivity behavior of the microemulsions. It was demonstrated that PPI favored the percolation process, shifting the Tp to lower temperatures. It seems that the dehydration of AOT polar heads and the occurrence of PPI in the AOT-water interface as a spacer reduced the rigidity of the interfacial film, enhancing the diffusion of sodium ions through the channels. The findings obtained here call for further exploration of the proposed microemulsion-dendrimer delivery vehicle. Additional experimental work, such as probing in vitro delivery of the dendrimer or drug-dendrimer complex from the microemulsions will examine the feasibility of the proposed carriers. References and Notes (1) Nanjwade, B. K.; Bechra, H. M.; Derkar, G. K.; Manvi, F. V.; Nanjwade, V. K. Eur. J. Pharm. Sci. 2009, 38, 185–196. (2) Wolinsky, J. B.; Grinstaff, M. W. AdV. Drug DeliVery ReV. 2008, 60, 1037–1055. (3) Gajbhiye, V.; Palanirajan, V. K.; Tekade, R. K.; Jain, N. K. J. Pharm. Pharmacol. 2009, 61, 989–1003.

Nir et al. (4) Tomalia, D. A.; Reyna, L. A.; Svenson, S. Biochem. Soc. Trans. 2007, 35, 61–67. (5) Venditto, V. J.; Regino, C. A. S.; Brechbiel, M. W. Mol. Pharm. 2005, 2, 302–311. (6) Jang, W. D.; Selim, K. M. K.; Lee, Ch. H.; Kang, I. K. Prog. Polym. Sci. 2009, 34, 1–23. (7) Cheng, Y.; Xu, T. Eur. J. Med. Chem. 2008, 43, 2291–2297. (8) Cheng, Y.; Wu, Q.; Li, Y.; Hu, J.; Xu, T. J. Phys. Chem. B 2009, 113, 8339–8346. (9) Klajnert, B.; Cortijo-Arellano, M.; Bryszewska, M.; Cladera, J. Biochem. Biophys. Res. Commun. 2006, 339, 577–582. (10) Witvrouw, M.; Fikkert, V.; Pluymers, W.; Matthews, B.; Mardel, K.; Schols, D.; Raff, J.; Debyser, Z.; De Clercq, E.; Holan, G.; Pannecouque, C. Mol. Pharmacol. 2000, 58, 1100–1108. (11) Klajnert, B.; Cortijo-Arellano, M.; Cladera, J.; Bryszewska, M. Biochem. Biophys. Res. Commun. 2006, 345, 21–28. (12) Heegaard, P. M. H.; Pedersen, H. G.; Flink, J.; Boas, U. FEBS Lett. 2004, 577, 127–133. (13) Strey, R. Colloid Polym. Sci. 1994, 272, 1005–1019. (14) Spernath, A.; Aserin, A. AdV. Colloid Interface Sci. 2006, 128130, 47–64. (15) Ezrahi, S.; Wachtel, E.; Aserin, A.; Garti, N. J. Colloid Interface Sci. 1997, 191, 277–290. (16) Kogan, A.; Garti, N. AdV. Colloid Interface Sci. 2006, 123-126, 369–385. (17) Feldman, Y.; Kozlovich, N.; Nir, I.; Garti, N.; Archipov, V.; Idiyatullin, Z.; Zuev, Y.; Fedotov, V. J. Phys. Chem. 1996, 100, 3745– 3748. (18) Chen, C. Z. S.; Cooper, S. L. Biomaterials 2002, 23, 3359–3368. (19) Scherrenberg, R.; Coussens, B.; van Vliet, P.; Edouard, G.; Brackman, J.; de Brabander, E.; Mortensen, K. Macromolecules 1998, 31, 456–461. (20) Kabanov, V. A.; Zezin, A. B.; Rogacheva, V. B.; Gulyaeva, Z. G.; Zansochova, M. F.; Joosten, J. G. H.; Brackman, J. Macromolecules 1998, 31, 5142–5144. (21) van Duijvenbode, R. C.; Borkovec, M.; Koper, G. J. M. Polymer 1998, 39, 2657–2664. (22) Senatra, D.; Lendinara, L.; Giri, M. G. Prog. Colloid Polym. Sci. 1991, 84, 122–128. (23) Yaghmur, A.; Aserin, A.; Tiunova, I.; Garti, N. J. Therm. Anal. Calorim. 2002, 69, 163–177. (24) Teubner, M.; Strey, R. J. Chem. Phys. 1987, 87, 3195–3200. (25) Senatra, D.; Ziparo, C.; Gambi, C. M. C.; Lanzi, L. J. Therm. Anal. Calorim. 2008, 2, 535–541. (26) Ezrahi, S.; Aserin, A.; Fanun, M.; Garti, N. Subzero Temperature Behavior of Water in Microemulsions. In Thermal BehaVior of Dispersed Systems; Surfactant Science Series; Garti, N., Ed.; Marcel Dekker, Inc.: New York, NY, U.S., 2001; Vol. 93, pp 59-120. (27) Moran, P. D.; Bowmaker, G. A.; Cooney, R. P.; Bartlett, J. R.; Woolfrey, J. L. Langmuir 1995, 11, 738–743. (28) Li, Q.; Weng, S. F.; Wu, J.; Zhou, N. F. J. Phys. Chem. B 1998, 102, 3168–3174. (29) Li, Q.; Li, T.; Wu, J. J. Phys. Chem. B 2000, 104, 9011–9016. (30) Nucci, N. V.; Vanderkooi, J. M. J. Phys. Chem. B 2005, 109, 18301–18309. (31) Brubach, J. B.; Mermet, A.; Filabozzi, A.; Gerschel, A.; Lairez, D.; Krafft, M. P.; Roy, P. J. Phys. Chem. B 2001, 105, 430–435. (32) Durantini, A. M.; Falcone, R. D.; Silber, J. J.; Correa, N. M. Chem. Phys. Chem. 2009, 10, 2034–2040. (33) Moran, P. D.; Bowmaker, G. A.; Cooney, R. P.; Bartlett, J. R.; Woolfrey, J. L. J. Mater. Chem. 1995, 5, 295–302. (34) Lowry, S. R.; Mauritz, K. A. J. Am. Chem. Soc. 1980, 102, 4665– 4667. (35) Mathew, C.; Patanjali, P. K.; Nabi, A.; Maitra, A. Colloids Surf. 1988, 30, 253–263. (36) Garcıá-Rıó, L.; Leis, J. R.; Mejuto, J. C.; Penã, M. E.; Iglesias, E. Langmuir 1994, 10, 1676–1683. (37) Paul, B. K.; Mitra, R. K. J. Colloid Interface Sci. 2006, 295, 230– 242. (38) Arias-Barros, S. I.; Cid, A.; Garcia-Rio, L.; Mejuto, J. C.; Morales, J. Colloid Polym. Sci. 2010, 288, 217–221.

JP108040Y

Solubilization of a Dendrimer into a Microemulsion - The Journal of

Recommend Documents