An Overview of the Catanionic Vesicle Technology

Technology Description:  Targeted Delivery Vesicles (TDVs) are vesicles formed from a mixture of cationic and anionic surfactants that form spontaneously in aqueous media.   The vesicles possess a number of unique physical and chemical properties that make them particularly attractive as materials for diagnostic and drug delivery applications when compared to liposomes.

The vesicles are

  •  unilamellar
  •  have a diameter of 150-200 ± 20 nm
  •  can be formulated as either cationic or anionic vesicles
  •  are thermodynamically stable at room temperature for months
  •  can be heated to 65 °C without decomposition
  •  will preferentially and efficiently encapsulate charged molecules
  •  can be formulated with biomolecules in the leaflet for targeting applications

Introduction

Vesicle-forming catanionic systems, i.e., mixtures of cationic and anionic surfactants, were first reported by Kaler et al. nearly two decades ago.(1) This report by Kaler led to research on the phase behavior of catanionic systems in general and vesicle forming mixtures in particular.  During the 1990’s, other vesicle-forming surfactant mixtures were discovered and characterized and theoretical models of spontaneous vesicle formation were developed.  However, until the reports of English,(2,3) the potential of catanionic vesicles for use in applications of molecular sequestration, as separations media and as targetable delivery systems had not been reported.  SD Nanosciences in collaboration with the English and DeShong laboratories at Wichita State University and the University of Maryland have developed several promising applications for catanionic vesicles, which are documented in peer-reviewed research articles (2-5) and in patent applications.(6,7)  SD Nanosciences has focused on molecular engineering approaches at the bilayer/water interface, a central theme in the processing of soft materials.  These interfacial engineering approaches produce functional interfacial properties of special interest for applications in biotechnology. We have demonstrated surface functionalization of catanionic vesicles using designed amphiphiles. (4,5)  The incorporation of surface functionalities into vesicles is readily accomplished through molecular self-assembly, driven by thermodynamic intermolecular forces.  These studies have developed robust tools based on catanionic surfactant vesicles for biotechnological applications.

Background

            The formation of vesicles from phospholipid molecules has been known since the first report fromBangham and Horne in 1964. (8)  In the 1970s the first reports of bilayer and vesicle formation from single-tailed surfactants emerged.9,10  Subsequent to those initial discoveries, vesicles have been used as cell membrane models and drug delivery vehicles.  In 1989 Kaler et al. reported the first example of spontaneous unilamellar vesicle formation from pairs of oppositely-charged single-tailed surfactants. (1) Their discovery led to reports of a number of other vesicle-forming systems composed of mixed surfactants that are now commonly referred to as catanionic mixtures and new systems continue to be reported.

Catanionic vesicles form spontaneously when unequal quantites of anionic and cationic single-tailed surfactants are dissoved in water (see Figure 1).  Figure 1B is a generic schematic of a dilute, aqueous, binary mixture of two oppositely charged surfactants and illustrates the salient features of vesicle-forming catanionic mixtures.  In addition to the precipitation zone near equimolarity, there are two lobe-shaped regions denoted V+ and V– in which spontaneous vesicle formation occurs.  These regions occur in many catanionic systems and indicate that stable catanionic vesicles always require one of the surfactants to be present in excess.  An example is the system consisting of cetyl trimethylammonium tosylate (CTAT), a 16-carbon-tailed cationic, and sodium dodecyl benzene sulfonate (SDBS), a 12-carbon-tailed anionic, where the lobes extend maximally at CTAT/SDBS weight ratios of 70/30 (V+) and 30/70 (V–).  The vesicle lobes fall between the precipitation zone and two regions where mixed micelles occur.  Unilamellar vesicles form spontaneously in mixtures within the lobes and their bilayers are rich in the major component.  This gives the bilayer a net charge that is responsible for the remarkable colloidal stability of these systems.  Spontaneously formed catanionic vesicles can be stable for years at room temperature. 

There are a number of reports aimed at developing a comprehensive understanding of spontaneous unilamellar vesicle formation in catanionic systems.  The central question is what are the directing parameters that lead to spontaneous curvature of a surfactant bilayer?  To address this question, one must consider factors such as tail-packing, head-group repulsion, electrostatic interactions and how these factors affect the bilayer bending modulus and free energy of vesicle formation.  An easily understood model for spontaneous vesicle formation in catanionic systems invokes the concept of non-ideal mixing between the two leaflets of a bilayer. (11)

Within this model, phase separation of the two surfactants in the inner and outer leaflets leads to a composition difference between the two leaflets of the bilayer.  Differing compositions of the two layers allows them to have oppositely signed curvatures, which is necessary for a net spontaneous curvature of the composite bilayer itself.  As with the theory of spontaneous curvature in a monolayer to form a micelle, concepts of asymmetry in the packing densities of the polar and non-polar regions of the surfactants can explain spontaneous curvature in the bilayer leaflets.  The formation of ion-pair complexes increases the packing densities of the surfactant head groups, while the presence of unpaired surfactant monomers will increase the average head group area due to electrostatic repulsions.  It follows that bending of a mixed surfactant bilayer would be favorable if the exterior monolayer had a larger fraction of the unpaired surfactant monomers, as suggested in the cartoon depiction in Figure 1A.  In this drawing, the outer leaflet holds a larger fraction of unpaired surfactant molecules and this induces positive curvature due to a low head-to-tail packing ratio.  The inner leaflet illustrates the opposite case. 

This model predicts that the external leaflet of catanionic vesicle bilayer has a larger mole fraction of the excess surfactant, than the interior leaflet.  Hence, these vesicles should have a high external surface charge and excellent colloidal stability, as is observed.  Moreover, the charged external bilayer should be highly attractive for charged solutes and counterions.  Indeed this has been observed with both low molecular weight molecules and DNA. (12-16)

Vesicle-forming catanionic mixtures are prepared from readily available and inexpensive surfactants.  This feature is viewed as a benefit when compared to vesicles formed from rare or expensive phospholipid components. In comparison with phospholipid vesicles, catanionic vesicles have a distinct advantage in terms of colloidal stability.  Spontaneously formed surfactant vesicles are naturally thermodynamically stable; whereas, vesicle preparations that require steps such as sonication or extrusion to break up larger lamellar structures are composed of kinetically trapped aggregates.  The degradation pathways for unilamellar phospholipid vesicles are well known and include fusion, rupture and spreading on hydrophilic surfaces.  Phospholipid vesicles are studied extensively as cell membrane models and are used commercially for drug delivery.  The superior stability and versatile physiochemical properties of catanionic vesicles puts them as a competitive alternative to conventional lipid-based vesicles.

To date, several catanionic vesicle applications have been demonstrated including their use as separations media,2 in preparation of various micro- and nanoparticles, (17) the formation of gels and networks,18 and for molecular encapsulation. (2,3)  These applications occur through the remarkable long-term stability and physicochemical versatility of catanionic vesicles and point toward the usefulness of surfactant vesicles for a range of biotechnological applications including drug delivery and diagnostics. 

SD Nanosciences has made recent advances in modifying the external surfaces of catanionic vesicles with carbohydrate moieties that enhance their potential for biotechnological applications with respect to targeting and diagnostics.  To date the company have focused our studies on catanionic vesicles in three areas: 

1)    electrostatic sequestration of solutes in vesicles and vesicle-based separations,

2)    surface functionalization of catanionic vesicles with carbohydrate moieties to interact with proteins in solution and at cell surfaces, and 

3) immobilization of vesicles using electrochemical deposition with chitosan for vesicle array formation.

Electrostatic Sequestration and Separation of Charged Solutes

Catanionic surfactant vesicles are able to capture and retain small, charged solutes.  The encapsulation efficiences of these vesicles are remarkably high. CTAT-rich vesicles (V+), from the CTAT/SDBS system were able to sequester the dye 5(6)-carboxyfluorescein (CF) to a much higher degree than uncharged phospholipid vesicles.  The phospholipid vesicles were formed from egg-yolk phosphatidylcholine (EYPC) by extrusion through a polycarbonate membrane and the catanionic vesicles were prepared from a 2:1 molar ratio of CTAT to SDBS (1% w/w total surfactant in water). The apparent encapsulation efficiency e measured by size exclusion chromatography (SEC) was 22% for V+ and only 1.6% for EYPC vesicles. The quantity e is referred to as the “apparent” encapsulation efficiency because we have shown that its measured value is nearly identical regardless of whether dye addition occurs during or after vesicle formation.  Adding the solute to pre-formed vesicles decreases the value of e by only about 30%.  The results indicate that molecular “encapsulation” by catanionic vesicles of oppositely charged solutes is due largely to adsorption of molecules to the vesicle exterior through electrostatic interactions with the excess surfactant present in the bilayer.

 We have measured e for several dyes and the drug doxorubicin, a cancer chemotherapeutic that has been delivered in a liposomal formulation. The values are shown in Table 1: the e values are always high and range from ca. 20-75% for oppositely charged dye and drug molecules. 

The release of sequestered dye (or drug) occurs slowly when compared to liposome vesicles.  Figure 2 compares the release profiles of the anionic dye CF in surfactant vesicles (t1/2 ≈ 80 days) and vesicles formed from neutral phospholipids (t1/2 ≈ 2 days).  Studies with the other dyes in Table 1 gave similar results.  In all cases studied, dye release was slow relative to neutral phospholipid vesicles.

The slow release profiles point to a strong dye/vesicle interaction and a slow desorption rate. An interesting way to take advantage of the strong adsorption and slow release characteristics is illustrated by a recent practical application of the vesicle technology.  We have used catanionic vesicles to separate mixtures of small molecules of similar molecular weight but opposite charge.  No special equipment is required other than a stir plate and a burette filled with Sephadex gel.  The results are shown in Figure 3. CTAT/SDBS vesicles were prepared in equimolar solutions of two oppositely charged dyes (CF and rhodamine 6G, R6G) and the normal method of separating vesicles from free dye through SEC was carried out.

 In Figure 3a, CTAT-rich V+ vesicles were added to a dye solution (0.5 mM total dye concentration) and SEC was performed.  UV-vis analysis showed that 31% of the anionic CF elutes with the vesicle band and no detectable R6G emerges with the vesicles.  In short, the V+ vesicles are able to selectively capture the anionic dye, and thereby separate it from the dye mixture.  The opposite behavior is observed for the same dye mixture in the presence of SDBS-rich V– vesicles (Figure 3b).  In this case, the vesicle band emerging from the SEC column contains 88% of the R6G, while the amount of CF in this band is negligible.  Thus, the anionic vesicles are able to bind and separate the cationic dye from the dye mixture.  This is the first demonstration of using surfactant vesicles as a means to separate ionic compounds.  The same experiments were conducted with a total dye concentration of 1.0 mM CF and R6G, and similar results were obtained.  Separation experiments using the anionic dye Lucifer Yellow, and the cationic drug doxorubicin, were also performed, and very efficient separation using vesicles was again observed, much like in Figure 3.

Surface Functionalization of Catanionic Vesicles for Biological Targeting

We have decorated the exterior of vesicles with carbohydrates using single-tailed glycoconjugates consisting of a glycan portion and an alkyl chain of varying tail length and with different carbohydrate headgroups. In collaboration with the DeShong at UMD, we synthesized a series of glycolipids, studied their incorporation into CTAT/SDBS vesicles and measured their interactions with sugar-binding proteins in solution.  Our interest in carbohydrate modification of vesicles stems from the ability to use carbohydrates for cell targeting with high specificity and from the growing interest in glycomics and the need for new tools in this area. Carbohydrates are ubiquitous in organisms and are involved in cell-cell recognition including the infectivity of pathogens, immune response and reproduction.   SDBS-rich V– vesicles were prepared that contained single-tailed glycolipids with octyl tails inserted into the bilayer.  The accessibility of the carbohydrate groups on the vesicle surface was gauged by several proprietary assays.  Multivalent interactions arising from binding to glycolipids plays an important role in the biology of cell surfaces and protein-carbohydrate interactions are facilitated by both the ligand density and spatial arrangements.  Our results show that surfactant vesicles can be used as scaffolds to present carbohydrates at the necessary density and spatial arrangement for facile and specific binding by biological systems and thus have applications in drug delivery targeting systems and vaccine development. 

Figure 4 shows binding studies in which we monitor the turbidity of the solution as a function of added lectin.  Lectin induced agglutination provides an excellent observable for monitoring both the agglutination onset as a function of added lectin and binding kinetics.  Initial rates of binding are good measures of the forward binding rate.  Figures 4A and B depict the binding characteristics of both glucose (●) and lactose (■) functionalized vesicles with concanavalin A (Con A) and peanut agglutinin (PNA).   Con A is a plant lectin derived from the Jack Bean (Canavalia ensiformis) and provides a well-known model system for investigating sugar-lectin interactions.    The results in Figures 4A and B clearly demonstrate that incorporation of glycolipids into the vesicle leaflet had occurred and that the carbohydrates are displayed on the vesicle surface and can be bound by lectins.  These observations indicate that carbohydrate-functionalized vesicles provide a convenient mode of cellular targeting in drug delivery applications.

Toxicology Studies of Catanionic Vesicles

If catanionic vesicles are going to be employed in drug delivery, it is critical that their toxicity be studied to ensure their safety.   As a demonstration of how molecular engineering of the vesicle interface can be used to steer biological interactions, we have quantified the viability of cells incubated in the presence of CTAT/SDBS vesicles in collaboration with Professor John Fisher in the Department of Bioengineering at UMD.  We monitored the cell viability of primary cell lines as a function of total surfactant concentration for bare CTAT/SDBS vesicles along with vesicles that bear glucose glycoconjugates.  In terms of vesicle toxicity, we found similar results to those presented by Kuo et al. for SDS/HTMAB vesicles.#  Figure 5 shows results obtained from experiments with calf chondrocytes using three different vesicle formulations.  Positively charged CTAT-rich vesicles (7:3 w/w CTAT to SDBS) caused the greatest degree of cell death over the largest concentration range.  CTAT-rich vesicles effectively target the negatively charged cell plasma membrane through non-specific electrostatic interactions delivering high dose of surfactants even at low concentrations.  SDBS-rich vesicles showed good cell viability when surfactant concentration was at or below 0.01 % (0.27 mM total surfactant).  The increased cell viability for SDBS-rich vesicles illustrates that non-specific interactions are diminished making these vesicles less toxic and good candidates for targeted delivery.  We tested SDBS-rich vesicles containing 0.3 mole fraction of C12-glucose. Glucose functionalized vesicles have a toxicity profile similar to that of positively-charged vesicles even though the vesicle charge is still highly negative (z potential = -50 mV).

Modification of the vesicle exterior with glycoconjugate induced a specific interaction with glucose receptors on the cell membrane and provided another indication that functionalized vesicles can be utilized for targeted vesicle delivery.   Moreover, this example shows that our surface modification methodology creates catanionic vesicles with surface displayed carbohydrates that are biologically active.

References

(1)       Kaler, E. W.; Murthy, A. K.; Rodriguez, B. E.; Zasadzinski, J. A. N. Science 1989, 245, 1371.

(2)       Danoff, E. J.; Wang, X.; Tung, S.-H.; Sinkov, N. A.; Kemme, A. M.; Raghavan, S. R.; English, D. S. Langmuir 2007, 23, 8965.

(3)       Wang, X.; Danoff, E. J.; Sinkov, N. A.; Lee, J.-H.; Raghavan, S. R.; English, D. S. Langmuir 2006, 22, 6461.

(4)       Park, J.; Rader, L. H.; Thomas, G. B.; Danoff, E. J.; English, D. S.; DeShong, P. Soft Matter 2008, 4, 1916.

(5)       Thomas, G. B.; Rader, L. H.; Park, J.; Abezgauz, L.; Danino, D.; DeShong, P.; English, D. S. J. Amer. Chem. Soc. 2009, In Press.

(6)       English, D. S.; DeShong, P.; Stein, D. C.; Lioi, S.; Park, J.; Danoff, E. J.; Thomas, G. B. Carbohydrate Functionalized Catanionic Surfactant Vesicles for Drug Delivery.; Application, P. I., Ed., 2008.

(7)       English, D. S.; Wang, X.; Danoff, E. J.; Sinkov, N. A.; Lee, J.-H.; Raghavan, S. R.; DeShong, P. Encapsulation and separation of charged organic solutes inside catanionic vesicles, 2008.

(8)       Bangham, A. D.; Horne, R. W. J. Mol. Biol. 1964, 8, 660.

(9)       Gebicki, J. M.; Hicks, M. Nature 1973, 243, 232.

(10)     Hargreaves, W. R.; Deamer, D. W. Biochemistry 1978, 17, 3759.

(11)     Safran, S. A.; Pincus, P.; Andelman, D. Science 1990, 248, 354.

(12)     Bonincontro, A.; La Mesa, C.; Proietti, C.; Risuleo, G. Biomacromolecules 2007, 8, 1824.

(13)     Dias, R. S.; Lindman, B.; Miguel, M. G. J. Phys. Chem. B 2002, 106, 12600.

(14)     Mel'nikov, S. M.; Dias, R.; Mel'nikova, Y. S.; Marques, E. F.; Miguel, M. G.; Lindman, B. FEBS Letters 1999, 453, 113.

(15)     Rosa, M.; Miguel, M. d. G.; Lindman, B. J Colloid Interf Sci 2007, 312, 87.

(16)     Rosa, M.; Moran, M. d. C.; Miguel, M. d. G.; Lindman, B. Colloids and Surfaces, A: Physicochemical and Engineering Aspects 2007, 301, 361.

(17)     McKelvey, C. A.; Kaler, E. W.; Zasadzinski, J. A.; Coldren, B.; Jung, H. T. Langmuir 2000, 16, 8285.

(18)     Payne, G. F.; Raghavan, S. R. Soft Matter 2007, 3, 521.