Novel Colloidal Microstructures of β-Escin and the Liposomal Components Cholesterol and DPPC

• Abstract
• Introduction
• Results
• Discussion
• Materials and Methods
• Materials
• Methods
• Langmuir film balance technique
• Preparation of saponin-containing dispersions
• Colorimetric quantification of dipalmitoylphosphatidylcholine
• High-performance liquid chromatography
• Critical micelle concentration
• Transmission electron microscopy
• References

Abstract

The discovery of immunostimulating complex formation by the saponin Quil A from the plant Quillaja saponaria with cholesterol and a phospholipid opened up new avenues for the development of drug delivery systems for vaccine application with additional adjuvant properties. In this study, β-escin, a monodesmosidic triterpene saponin from horse chestnut, was investigated in terms of its interaction with liposomal components (cholesterol, dipalmitoylphosphatidylcholine) by Langmuir film balance studies and with regard to particle formation visualized by transmission electron microscopy. A strong interaction of β-escin with cholesterol was observed by Langmuir isotherms due to the intercalation of the saponin into the monolayer, whereas no interaction occurred with dipalmitoylphosphatidylcholine. Transmission electron microscopy studies also confirmed the strong interaction of β-escin with cholesterol. In aqueous pseudo-ternary systems (β-escin, dipalmitoylphosphatidylcholine, cholesterol) and in pseudo-binary systems (β-escin, cholesterol), new colloidal structures built up from ring-like and worm-like subunits were observed with a size of about 100 – 200 nm. These colloidal structures are formed in pseudo-binary systems by aggregation of the subunits, whereas in pseudo-ternary systems, they are formed among others by attacking the liposomal membrane. The rehydration of the liposomal dispersions in NANOpure water or Tris buffer pH 7.4 (140 mM) resulted in the same particle formation. In contrast, the sequence of the dispersionsʼ production process affected the particle formation. Unless adding the saponin to the other components from the beginning, just a liposomal dispersion was formed without any colloidal aggregates of the subunits mentioned above.

Introduction

Saponins are becoming more and more into focus, as many pharmacological and biological properties have been found in recent years [1]. They occur in many plants and one of their characteristic features is their foam forming activity in aqueous solution because of their amphiphilic character. Saponins are subdivided into steroid (C₂₇) and triterpene (C₃₀) glycosides attached with one (monodesmosidic) or two (bidesmosidic) sugar chains of different sugar units, conferring the molecule an amphiphilic character [2].

Quillaja saponin is a triterpene saponin from the bark of Quillaja saponaria Molina (Quillajaceae) and has widely been studied in terms of vaccine development due to its highly immunoadjuvant effect [3]. Quil A, a semi-purified fraction of Quillaja saponin, induces both the humoral and cellular immune response, which is a major advantage over conventional adjuvants [4]. Chemically, Quil A belongs to the class of bidesmosidic triterpene saponins and forms micelles in aqueous solution above a CMC [5]. In the presence of phospholipid and cholesterol, Quil A forms a cage-like structure in the size range of 40 – 100 nm built up from ring-like subunits holding together due to hydrophobic interactions (Van-der-Waals forces) and hydrogen bonds [6], [7], [8]. The major reason for the formation of colloidal structures, observed from Langmuir film balance experiments, was the hydrophobic interaction between cholesterol and Quil A [9]. Due to immunostimulating properties, the cage-like structure is known as an ISCOM matrix. For vaccine development, antigens are able to be incorporated into ISCOM matrices. ISCOM matrices represent promising drug delivery systems with high adjuvant effects, resulting in a reduction of the loaded antigen concentration. They are particularly attractive for subunit vaccines that cause only a weak immune response.

Due to the structural characteristics of saponins and their high affinity to cholesterol, we suggested in a previous review that other saponins should also form colloidal structures with liposomal components [10]. Recently, this was confirmed for further saponins in terms of ISCOM formation and adjuvant properties [11]. Saponins from Allochrusa gypsophiloides (Regel) Schischk. (Caryophyllaceae), L. (Caryophyllaceae), and Gypsophila paniculata L. (Caryophyllaceae) formed ISCOM-like structures upon the addition of lipids (egg phospholipids and cholesterol), which proved to have an immunostimulatory effect [11]. Therefore, it is interesting to investigate other promising saponins regarding their structure formation with liposomal components such as cholesterol and DPPC. A structurally different saponin compared to Quil A is the triterpene saponin β-escin from the seeds of horse chestnut (Aesculus hippocastanum L., Sapindaceae) [12], [13]. Both saponins share a similar aglycon, but there are distinct differences in the side chains ([Fig. 1]). β-Escin is a monodesmosidic saponin attached with one sugar chain composed of three sugar units, while Quil A is a bidesmosidic saponin attached with two sugar chains. The missing sugar chain in the β-escin molecule makes it less water soluble. β-Escin is established as a drug for the treatment of vein weakness. It has anti-edematous, anti-inflammatory, and venotonic properties [12].

The aim of this study was to investigate the interaction of β-escin with liposomal components such as cholesterol and DPPC by Langmuir film balance experiments. This method determines the changes in surface pressure of lipid monolayers (as model membranes) onto different subphases during constant compression of the monolayer to detect interactions between the saponin molecules in the subphase and the lipid molecules in the surface. In order to investigate colloidal structure formation of β-escin with the liposomal components cholesterol and DPPC, TEM was used for the visualization of nanometer-sized colloidal aggregates.

Results

Langmuir isotherms (π/A isotherms) of β-escin and Quil A are shown in [Figs. 2] and [3], while the corresponding collapse points are summarized in [Table 1]. Different monolayer compositions of cholesterol and DPPC were investigated on an aqueous subphase and compared to a saponin-containing subphase. The collapse point for a pure cholesterol monolayer on an aqueous subphase was determined at 36.42 Ų ([Figs. 2 A] and [3 A]), which agrees with the literature [14]. The isotherm on a β-escin subphase (10⁻³ mg/mL) resulted in a shift to a larger molecular area of 42.29 Ų ([Fig. 2 A]). The area at the collapse pressure increased by 5.87 Ų upon the addition of β-escin into the subphase. For a mixed DPPC/cholesterol monolayer (3 : 1, w/w), similar observations were made ([Fig. 2 C]). The DPPC molecules require more space in the monolayer, resulting in a higher molecular area at the collapse point of a DPPC/cholesterol mixed monolayer on an aqueous subphase (42.60 Ų) compared to a pure cholesterol monolayer. The exchange of the aqueous subphase with a β-escin-containing subphase resulted in a shift to a larger molecular area of 45.33 Ų at the collapse point. The shift of the isotherm (2.73 Ų) after replacement of the aqueous subphase with a β-escin-containing one was lower compared to a pure cholesterol monolayer ([Fig. 2 A, C]). For a pure DPPC monolayer, different isotherms were observed ([Fig. 2 B]). A pure DPPC monolayer on an aqueous subphase collapsed at the highest molecular area of 45.82 Ų. Upon addition of the saponin (10⁻³ mg/mL) into the subphase, no shift of the isotherm to a larger molecule area was observed; the curves overlapped. Similar results were observed for the saponin Quil A at a concentration of 10⁻⁴ mg/mL in the subphase. These data correspond to previous results from our group with Quil A and phosphatidylcholine (PC) [9] while in the present study, DPPC was studied, suggesting no interaction with the saponin.

π/t Measurements of lipid monolayers on a β-escin-containing subphase are shown in [Fig. 4], while the initial and final surface pressure are listed in [Table 2]. The π/t measurments were recorded over a period of 10 000 s. The surface pressure of a pure cholesterol monolayer on a β-escin-containing subphase with a concentration of 10⁻³ mg/mL increased over time ([Fig. 4 A]). The less compressed monolayer (lower curve) showed a greater increase in surface pressure over time compared to the more compressed monolayer (upper curve) due to the lower packing density of the molecules. At given intervals, both diagrams approached almost constant surface pressures, yet at different levels. For a mixed DPPC/cholesterol monolayer (3 : 1, w/w) and a pure DPPC monolayer, different results were observed ([Fig. 4 B, C]). The less compressed monolayers of both DPPC alone and DPPC/cholesterol on β-escin-containing subphases (lower curves) showed only a slight increase in surface pressure over time. The surface pressures of the more compressed monolayers (upper curves) remained approximately constant over a period of 10 000 s with the exception of a slight drop in surface pressure after starting the experiment in the case of the mixed DPPC/cholesterol monolayer.

The aqueous β-escin-containing subphase of the mixed DPPC/cholesterol monolayer was investigated by TEM. Two micrographs ([Fig. 5 A, B]) show liposomes with a budding structure at their surfaces (see arrows). The latter structure consists of ring-like subunits self-assembling into almost circular aggregates. Obviously, molecules (DPPC, cholesterol) from the monolayer moved into the subphase to form liposomes, which were attacked by saponin molecules from the subphase with formation of the specific structure.

In order to visualize the influence of β-escin on lipid components, further TEM (transmission electron microscope) images of pseudo-binary and pseudo-ternary systems were examined ([Figs. 6] and [7]). [Fig. 6 A] shows an aqueous β-escin solution (0.1 mg/mL). Above a concentration of 0.065 mg/mL, β-escin forms micelles in aqueous solution. The electron micrograph reveals circular and oval-shaped micelles. The latter have an average length of about 20 nm, similar to the diameter of the circular structures. In [Fig. 6 B, C], a pseudo-ternary system of DPPC, cholesterol, and β-escin in water is shown, directly after preparation with the film method ([Fig. 6 B]), and after 2 months of storage ([Fig. 6 C]). After 2 weeks of storage, the sample was also examined and showed the same appearance as after 2 months of storage (data not shown). The aqueous pseudo-ternary mixture of DPPC, cholesterol, and β-escin forms colloidal structures of heterogeneous size of about 150 nm with approximately spherical geometry ([Figs. 6 B, C] and [7 A]). These structures are built from worm-like and ring-like subunits (as indicated by the arrows in [Fig. 7 A]). [Fig. 7 B] shows the formation of these colloidal structures as budding from a liposomal membrane. The liposomal membrane is disrupted at the site of colloidal structure formation. The colloidal structures are to be observed individually or in combination with several ones. In addition to the formed novel structures, oval-shaped micelles are visible (as indicated by an arrow in [Fig. 6 B]). The storage samples remained stable over the respective storage period and present the same spherical colloidal structures ([Fig. 6 C]). A pseudo-binary system of DPPC and β-escin in water forms different structures, as can be seen in [Fig. 6 D]. Oval subunits with an average length of 15 nm are stacked together in a worm-like arrangement, forming chains of different lengths with cavities between the oval subunits. In addition, circular structures were also observed as in the micellar solution. The oval subunits are also observed in the pseudo-ternary mixtures with a slightly larger average diameter of 20 nm (as indicated by an arrow in [Fig. 6 B]). In comparison to the pseudo-ternary mixture, no spherical colloidal structures of larger dimensions are visible. An aqueous system of cholesterol and β-escin forms colloidal structures with spherical geometry built from worm-like and ring-like subunits in the same way as the pseudo-ternary mixture ([Fig. 6 E]). In contrast to the pseudo-ternary mixture, no oval subunits are observed in the surroundings of those colloidal structures.

Due to the low solubility of pure β-escin in water and its higher solubility in Tris buffer of pH 7.4 (140 mM), the preparation of the dispersions via the film method and subsequent hydration was also done with Tris buffer. Tris buffer does not modify the formation of colloidal structures composed of DPPC, cholesterol, and β-escin as shown in [Fig. 6 F – H]. The same structures are displayed as in the aqueous dispersions. All the systems were repeated with different ratios of all the components confirmed by HPLC (β-escin and cholesterol) and the Stewart assay (DPPC) ([Table 3]). The same structures were found as in [Fig. 6 A – H] (data not shown). However, significantly fewer colloidal structures were observed in those pseudo-ternary and pseudo-binary systems containing β-escin and cholesterol, as the cholesterol concentration was very low.

In addition, the film method for the preparation of the pseudo-ternary system was slightly modified to determine its influence on colloidal particle formation. An aqueous β-escin solution was added for rehydration after film formation of cholesterol and DPPC. In this case, a liposomal dispersion was observed without any colloidal structures, as presented in the previous electron micrographs (data not shown). Obviously, contact between the components was not sufficient for the formation of novel colloidal structures and/or more time was needed.

Discussion

With regard to the affinity of β-escin to liposomal components, the Langmuir film balance technique demonstrates a strong interaction of β-escin with cholesterol. The results of the isotherms clearly show that β-escin enters the cholesterol monolayer from the subphase, as seen from a significant shift of the isotherm to the increased area. The intercalation of β-escin molecules in the monolayer increases the space between cholesterol molecules, resulting in a collapse point at a larger molecular area. A strong shift of the isotherm to larger molecular areas indicates strong interactions between the molecules of the subphase and the molecules in the monolayers. The intercalation of β-escin molecules in a mixed monolayer of DPPC and cholesterol is lower compared to a pure cholesterol monolayer, whereas in a pure DPPC monolayer, no intercalation of saponin molecules is detectable. The saponin molecules remain in the subphase and do not deposit between the DPPC molecules of the monolayer. The interaction between β-escin and the mixed monolayer of DPPC and cholesterol is likely due to the high affinity of β-escin towards cholesterol. The β-escin molecules will most likely intercalate between the cholesterol molecules of the mixed monolayer only.

Similar results were also achieved with the π/t measurements. Prior to the measurement, the monolayer on an aqueous subphase was precompressed to a given surface pressure. In a second step, the subphase was replaced by a saponin-containing subphase and the measurement of surface pressure versus time was started. A strong interaction between the β-escin molecules of the subphase and the cholesterol molecules in the monolayer was again demonstrated. At a low precompression of about 7.7 mN/m, and thus a large area per cholesterol molecule, in the monolayer, β-escin molecules are allowed to migrate into the monolayer, and the surface pressure increases up to 31 mN/m. Even starting at a precompression of the monolayer of 25 mN/m, which means the space per cholesterol molecule is drastically reduced, causes the β-escin molecules to intercalate in the monolayer over time. The final surface pressure approaches 36.83 mN/m at equilibrium, which means that the mixed monolayer of cholesterol and saponin is even more densely packed than that at an initially lower precompression. β-Escin showed no interaction with a pure DPPC monolayer over time, which also confirms the results of the π/A isotherms. The slight increase of the less compressed monolayer is likely due to a rather limited penetration of β-escin molecules into the monolayer due to a given mobility of the molecules within the monolayer. With regard to time-dependent measurements, β-escin showed no interaction with a mixed monolayer of DPPC and cholesterol, which contradicts the results of the π/A isotherm. Due to the controversial results, the β-escin-containing subphase of the mixed DPPC/cholesterol monolayer was investigated using TEM. Conspicuous structures on liposomal membranes were observed. These ring-like subunits self-assembling into almost circular aggregates indicate an interaction between components of the monolayer and the β-escin molecules in the subphase. Possibly, an exchange of molecules (DPPC, cholesterol) from the monolayer and β-escin molecules from the subphase occurred, which did not result in a noticeable change in surface pressure. In the subphase, the molecules of the monolayer formed liposomes. Due to the high affinity of β-escin to cholesterol, the saponin molecules attacked the liposomal membrane, resulting in associates of ring-like subunits.

The TEM images of pseudo-binary and pseudo-ternary systems also confirm the results of the Langmuir studies with regard to the strong affinity of β-escin to cholesterol. Novel colloidal structures are visible in aqueous pseudo-binary (cholesterol, β-escin) and pseudo-ternary systems (cholesterol, DPPC, β-escin). In addition to these colloidal structures, ring-like, worm-like as well as oval subunits were observed, indicating that such structures are involved in the formation of these novel structures, as suggested by Demana et al. for ISCOM formation [15]. These novel structures have a diameter between 100 – 200 nm and are therefore slightly larger than the known ISCOMs. The aggregation of several subunits to colloidal structures may be due to the fact that β-escin does not contain a charged sugar group compared to Quil A, which protects against aggregation by repulsion [7]. Interestingly, these colloidal structures were also formed in the absence of DPPC in contrast to the Quil A-containing ISCOM matrices, which are formed only in the presence of all three components (phospholipid, cholesterol, Quil A), as described in the literature [7], [16]. No further structures were observed in the vicinity of the colloidal aggregates from β-escin and cholesterol, suggesting complete solubilization of cholesterol by β-escin. As early as 1962, Bangham and Horne investigated the formation of a hexagonal lattice consisting of ring-shaped subunits from cholesterol and a saponin (of unknown origin) [17]. Thirty years later, Kersten et al. and other research groups observed the formation of ring-shaped micelles in an aqueous pseudo-binary mixture of cholesterol and Quil A [7], [18]. Our investigations have shown that these ring-shaped subunits aggregate into almost spherical associates. DPPC seems not to be essential for the formation of these novel structures, which is in line with the Langmuir experiments. The structure formation in pseudo-binary systems occurs by aggregation of ring-shaped subunits, whereas in ternary mixtures it also occurs by attacking the liposomal membrane, which was also demonstrated in the micrograph of the β-escin-containing subphase of a mixed DPPC/cholesterol monolayer.

Another interesting observation is the structure of the β-escin micelles in aqueous solution. The agglomerates of the oval-shaped micelles seem to be laterally aligned flat discs. These are also found in the pseudo-binary systems of β-escin and DPPC, with the difference that DPPC has an influence on the formation of long worm-like chains.

In this paper, novel structures built up from β-escin, cholesterol, and DPPC were presented. Cholesterol seems to be an essential component for the structure formation. These colloidal structures may be promising for the use as drug delivery systems. A limiting factor could be the hemolytic property of saponins, which has also been described for escin, the crude extraxt of the seeds of horse chestnut [12]. This may correlate with the high affinity to cholesterol. But maybe the complexation of the saponin with cholesterol and DPPC protects the erythrocytes from hemolysis, which has not been examined yet.

Materials and Methods

Materials

Cholesterol (Chol, ≥ 99% purity), purified Quillaja saponin (European Pharmacopoeia Reference Standard), and ammonium thiocyanate were purchased from Sigma-Aldrich. The triterpene saponin β-escin (98.7% purity) was obtained from MP Biomedicals. 1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC, 99% purity) was kindly donated by Lipoid. Ferric chloride hexahydrate was purchased from Roth. Water was purified by ultrafiltration (Millipore). Organic solvents for the preparation of saponin-containing dispersions, for cleaning the Langmuir trough, and for HPLC experiments were of analytical grade. All the substances were used as received.

Methods

Langmuir film balance technique

For surface pressure (π) measurements, a NIMA 611 Langmuir trough equipped with a film balance placed on an antivibration table was used. The Langmuir trough coated with inert polytetrafluorethylene contained the aqueous subphase. The system was equipped with two barriers (one movable, one static). Langmuir trough measurements were conducted by using the Wilhelmy plate method. The Wilhelmy plates were made from chromatography paper (Whatman CHR1 chromatography paper, perimeter 20.6 mm) of 10 mm width and 20 mm length. The Wilhelmy plate was attached to an S-ring connected to the pressure sensor. The lipids were dripped as solutions (chloroform) onto the subphase (either pure water or the aqueous saponin solution) using a microliter syringe with a reproducibility adaptor (Hamilton) and were left for 15 min to allow evaporation of the solvent and formation of the lipid film. The temperature was controlled and kept constant at 20 °C for all measurements. With the movable barrier, the compression isotherms were recorded by continuous compression of the monolayer (π/A isotherms). The experiment ended at the collapse point where the molecules were closely packed, overlapped, and the monolayer finally collapses.

For time-dependent measurements (π versus t), the monolayer on an aqueous subphase was first precompressed to a given pressure. The compression speed of the barrier was then set to zero. Subsequently, 5 mL of the aqueous subphase was removed with a curved cannula behind the barrier from below the monolayer and then 5 mL of a concentrated aqueous saponin solution was slowly injected under the monolayer to obtain the required concentration adjustment. After replacing the aqueous subphase with a β-escin subphase, the surface pressure development was followed over 10 000 s. At the end of the measurement, 2 mL of the saponin-containing subphase of the mixed DPPC/cholesterol monolayer was carefully removed with a cannula, as described above, and subsequently examined via TEM.

After each measurement, the trough and the barriers were thoroughly cleaned by wiping with isopropanol and rinsing with ultrapure water. The concentration for β-escin in the subphase was 10⁻³ mg/mL. Due to the high interaction of Quil A with the monolayer, a lower concentration (10⁻⁴ mg/mL) of Quil A was dissolved in the subphase to obtain reliable isotherms.

Preparation of saponin-containing dispersions

Saponin-loaded (β-escin) liposomal dispersions were prepared by the lipid film method according to Bangham et al. with slight modification [4]. The liposomal components (DPPC, cholesterol) and the saponin were dissolved in methanol to obtain a clear lipid solution. The organic solvent was then removed under vacuum at 45 °C by a rotary evaporator. The dried lipid film on the wall of the round bottom flask was rehydrated in NANOpure water or in Tris buffer pH 7.4 (140 mM) and then agitated to obtain a homogeneous dispersion [17]. After hydration of the film, the dispersion appeared turbid, and small particles were visible. Therefore, a filtration step was carried out through a PVDF membrane with a pore size of 0.22 µm (Carl Roth). The samples for storage were again filtered through a sterile filter with a pore size of 0.2 µm to prevent microbial growth. Subsequent to filtration, the composition of the systems was quantified and is presented in [Table 3] as “real” in comparison to “original” composition obtained from weighing. In order to determine a potential influence of the preparation procedure on the sample structure, an aqueous β-escin solution was used for rehydration of the lipid film of cholesterol and DPPC.

Colorimetric quantification of dipalmitoylphosphatidylcholine

For quantification of DPPC in pseudo-binary and pseudo-ternary systems after filtration through a PVDF membrane, a colorimetric method, the Stewart Assay, was used [19]. The real concentrations of DPPC in different formulations are shown in [Table 3]. This assay is based on a complex formation of the phospholipid with a color reagent, ammonium ferrothiocyanate. The color reagent is formed by dissolving ferric chloride hexahydrate (FeCl₃×6H₂O) and ammonium thiocyanate (NH₄SCN) in distilled water.

The calibration was performed in a concentration range of 0 – 0.5 mg/mL. A stock solution of DPPC in chloroform (1.0 mg/mL) was prepared, from which the corresponding concentrations were diluted so that each sample had a total volume of 2 mL of chloroform. The solutions were filled into centrifuge tubes and were coated with 2 mL of ammonium ferrothiocyanate. The two-phase system was vigorously mixed with a IKA MS 1 Minishaker for 60 s. The system was then centrifuged (104 × g, 5 minutes) for complete phase separation. The lower chloroform phase was removed with a Pasteur pipette and its optical density was then examined at λ 485 nm with an infinite M Plex Tecan microplate reader.

High-performance liquid chromatography

The quantification of cholesterol and β-escin in filtered dispersions was performed by reversed-phase HPLC using a Waters HPLC system (Waters). The RP column (GromHypersil ODS, 250 mm × 4 mm I. D., 5 µm) was kept at 30 °C in a thermostat. Isocratic elution was carried out with a mobile phase consisting of methanol/water 95/5 (v/v) at a flow rate of 1.0 mL/min. The mobile phase was degassed in an ultrasonic bath before use. The UV detector was set to 210 nm. Samples for calibration were dissolved in pure methanol and the injection volume for each sample was 50 µL. The real concentrations of cholesterol and β-escin in different formulations are shown in [Table 3].

Critical micelle concentration

The CMC of β-escin was determined by using a tensiometer Krüss K 100. A platinum Wilhelmy plate was used. In order to exclude impurities, the plate was cleaned with isopropanol and then annealed. After positioning the plate on the surface of the liquid, the force due to wetting was measured, which is proportional to the surface tension. The temperature was kept constant at 25 °C. Next, 20 mL of demineralized water was filled into a glass vessel with a capacity of 43.5 mL. The aqueous saponin solution of β-escin was gradually added (dropwise). In order to achieve a homogeneous distribution, the solution was stirred for 1 min after each addition and further equilibrated for 6 min before starting the measurement. Measurements were taken in duplicate. As soon as the surface tension remained almost constant and did not decrease further upon the addition of the saponin solution, the respective concentration was taken as the CMC.

Transmission electron microscopy

For sample preparation, a carbon-coated mica was added to the sample drop to float off the carbon support film. After 1 min, the carbon film with the absorbed sample was taken up by a copper grid (300 mesh), washed once in distilled water, and negatively stained with 4% (w/v) aqueous uranyl acetate. Samples were examined with a Zeiss Libra 120 Plus at an acceleration voltage of 120 kV.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgements

The supply of DPPC from Lipoid GmbH (Ludwigshafen, Germany) is gratefully acknowlegded.

• References

• 1 Sparg SG, Light ME, Van Staden J. Biological activities and distribution of plant saponins. J Ethnopharmacol 2004; 94: 219-243
  CrossrefPubMedSearch in Google Scholar
• 2 Vincken JP, Heng L, de Groot A, Gruppen H. Saponins, classification and occurrence in the plant kingdom. Phytochemistry 2007; 68: 275-297
  CrossrefPubMedSearch in Google Scholar
• 3 Sanders MT, Brown LE, Deliyannis G, Pearse MJ. ISCOM-based vaccines: the second decade. Immunol Cell Biol 2005; 83: 119-128
  CrossrefPubMedSearch in Google Scholar
• 4 Bangham AD, Standish MM, Watkins JC. Diffusion of univalent ions across the lamellae of swollen phospholipids. J Mol Biol 1965; 13: 238-252
  CrossrefPubMedSearch in Google Scholar
• 5 Morein B, Sundquist B, Höglund S, Dalsgaard K, Osterhaus ADME. Iscom, a novel structure for antigenic presentation of membrane proteins from enveloped viruses. Nature 1984; 308: 457-460
  CrossrefPubMedSearch in Google Scholar
• 6 Morein B, Lövgren K, Höglund S, Sundquist B. The ISCOM: an immunostimulating complex. Immunol Today 1987; 8: 333-338
  CrossrefPubMedSearch in Google Scholar
• 7 Kersten GF, Spiekstra A, Beuvery EC, Crommelin DJ. On the structure of immune-stimulating saponin-lipid complexes (iscoms). Biochim Biophys Acta 1991; 1062: 165-171
  CrossrefPubMedSearch in Google Scholar
• 8 Özel M, Höglund S, Gelderblom HR, Morein B. Quaternary structure of the immunostimulating complex (iscom). J Ultrastruct Mol Struct Res 1989; 102: 240-248
  CrossrefPubMedSearch in Google Scholar
• 9 Paepenmüller T, Müller-Goymann CC. Influence of Quil A on liposomal membranes. Int J Pharm 2014; 475: 138-146
  CrossrefPubMedSearch in Google Scholar
• 10 de Groot C, Müller-Goymann CC. Saponin interactions with model membrane systems – langmuir monolayer studies, hemolysis and formation of ISCOMs. Planta Med 2016; 82: 1496-1512
  Thieme ConnectPubMedSearch in Google Scholar
• 11 Turmagambetova AS, Alexyuk PG, Bogoyavlenskiy AP, Zaitseva IA, Omirtaeva ES, Alexyuk MS, Berezin VE. Adjuvant activity of saponins from Kazakhstani plants on the immune responses to subunit influenza vaccine. Arch Virol 2017; 126: 1-10
  PubMedSearch in Google Scholar
• 12 Sirtori CR. Aescin: pharmacology, pharmacokinetics and therapeutic profile. Pharmacol Res 2011; 44: 183-193
  PubMedSearch in Google Scholar
• 13 Wu XJ, Zhang ML, Cui XY, Fawcett JP, Gu JK. Comparative pharmacokinetics and the bioavailability of escin Ib and isoescin Ib following the administration of escin, pure escin Ib and isoescin Ib in rats. J Ethnopharmacol 2014; 151: 839-845
  CrossrefPubMedSearch in Google Scholar
• 14 Demel RA, Bruckdorfer KR, Van Deenen LLM. Structural requirements of sterols for the interaction with lecithin at the air-water interface. Biochim Biophys Acta 1972; 255: 311-320
  CrossrefPubMedSearch in Google Scholar
• 15 Demana PH, Davies NM, Vosgerau U, Rades T. Pseudo-ternary phase diagrams of aqueous mixtures of Quil A, cholesterol and phospholipid prepared by the lipid-film hydration method. Int J Pharm 2004; 270: 229-239
  CrossrefPubMedSearch in Google Scholar
• 16 Copland MJ, Rades T, Davies NM. Hydration of lipid films with an aqueous solution of Quil A: a simple method for the preparation of immune-stimulating complexes. Int J Pharm 2000; 196: 135-139
  CrossrefPubMedSearch in Google Scholar
• 17 Bangham AD, Horne RW. Action of saponin on biological cell membranes. Nature 1962; 196: 952-953
  CrossrefPubMedSearch in Google Scholar
• 18 Demana PH, Davies NM, Berger B, Vosgerau U, Rades T. A comparison of pseudo-ternary diagrams of aqueous mixtures of Quil A, cholesterol and phospholipid prepared by lipid-film hydration and dialysis. J Pharm Pharmacol 2004; 56: 573-580
  CrossrefPubMedSearch in Google Scholar
• 19 Stewart JCM. Colorimetric determination of phospholipids with ammonium ferrothiocyanate. Anal Biochem 1980; 104: 10-14
  CrossrefPubMedSearch in Google Scholar

Correspondence

• References

• 1 Sparg SG, Light ME, Van Staden J. Biological activities and distribution of plant saponins. J Ethnopharmacol 2004; 94: 219-243
  CrossrefPubMedSearch in Google Scholar
• 2 Vincken JP, Heng L, de Groot A, Gruppen H. Saponins, classification and occurrence in the plant kingdom. Phytochemistry 2007; 68: 275-297
  CrossrefPubMedSearch in Google Scholar
• 3 Sanders MT, Brown LE, Deliyannis G, Pearse MJ. ISCOM-based vaccines: the second decade. Immunol Cell Biol 2005; 83: 119-128
  CrossrefPubMedSearch in Google Scholar
• 4 Bangham AD, Standish MM, Watkins JC. Diffusion of univalent ions across the lamellae of swollen phospholipids. J Mol Biol 1965; 13: 238-252
  CrossrefPubMedSearch in Google Scholar
• 5 Morein B, Sundquist B, Höglund S, Dalsgaard K, Osterhaus ADME. Iscom, a novel structure for antigenic presentation of membrane proteins from enveloped viruses. Nature 1984; 308: 457-460
  CrossrefPubMedSearch in Google Scholar
• 6 Morein B, Lövgren K, Höglund S, Sundquist B. The ISCOM: an immunostimulating complex. Immunol Today 1987; 8: 333-338
  CrossrefPubMedSearch in Google Scholar
• 7 Kersten GF, Spiekstra A, Beuvery EC, Crommelin DJ. On the structure of immune-stimulating saponin-lipid complexes (iscoms). Biochim Biophys Acta 1991; 1062: 165-171
  CrossrefPubMedSearch in Google Scholar
• 8 Özel M, Höglund S, Gelderblom HR, Morein B. Quaternary structure of the immunostimulating complex (iscom). J Ultrastruct Mol Struct Res 1989; 102: 240-248
  CrossrefPubMedSearch in Google Scholar
• 9 Paepenmüller T, Müller-Goymann CC. Influence of Quil A on liposomal membranes. Int J Pharm 2014; 475: 138-146
  CrossrefPubMedSearch in Google Scholar
• 10 de Groot C, Müller-Goymann CC. Saponin interactions with model membrane systems – langmuir monolayer studies, hemolysis and formation of ISCOMs. Planta Med 2016; 82: 1496-1512
  Thieme ConnectPubMedSearch in Google Scholar
• 11 Turmagambetova AS, Alexyuk PG, Bogoyavlenskiy AP, Zaitseva IA, Omirtaeva ES, Alexyuk MS, Berezin VE. Adjuvant activity of saponins from Kazakhstani plants on the immune responses to subunit influenza vaccine. Arch Virol 2017; 126: 1-10
  PubMedSearch in Google Scholar
• 12 Sirtori CR. Aescin: pharmacology, pharmacokinetics and therapeutic profile. Pharmacol Res 2011; 44: 183-193
  PubMedSearch in Google Scholar
• 13 Wu XJ, Zhang ML, Cui XY, Fawcett JP, Gu JK. Comparative pharmacokinetics and the bioavailability of escin Ib and isoescin Ib following the administration of escin, pure escin Ib and isoescin Ib in rats. J Ethnopharmacol 2014; 151: 839-845
  CrossrefPubMedSearch in Google Scholar
• 14 Demel RA, Bruckdorfer KR, Van Deenen LLM. Structural requirements of sterols for the interaction with lecithin at the air-water interface. Biochim Biophys Acta 1972; 255: 311-320
  CrossrefPubMedSearch in Google Scholar
• 15 Demana PH, Davies NM, Vosgerau U, Rades T. Pseudo-ternary phase diagrams of aqueous mixtures of Quil A, cholesterol and phospholipid prepared by the lipid-film hydration method. Int J Pharm 2004; 270: 229-239
  CrossrefPubMedSearch in Google Scholar
• 16 Copland MJ, Rades T, Davies NM. Hydration of lipid films with an aqueous solution of Quil A: a simple method for the preparation of immune-stimulating complexes. Int J Pharm 2000; 196: 135-139
  CrossrefPubMedSearch in Google Scholar
• 17 Bangham AD, Horne RW. Action of saponin on biological cell membranes. Nature 1962; 196: 952-953
  CrossrefPubMedSearch in Google Scholar
• 18 Demana PH, Davies NM, Berger B, Vosgerau U, Rades T. A comparison of pseudo-ternary diagrams of aqueous mixtures of Quil A, cholesterol and phospholipid prepared by lipid-film hydration and dialysis. J Pharm Pharmacol 2004; 56: 573-580
  CrossrefPubMedSearch in Google Scholar
• 19 Stewart JCM. Colorimetric determination of phospholipids with ammonium ferrothiocyanate. Anal Biochem 1980; 104: 10-14
  CrossrefPubMedSearch in Google Scholar