Study of cholesterol’s effect on the properties of catanionic vesicular systems: comparison of light-scattering results with ultrasonic and fluorescence spectroscopy

Study of cholesterol’s effect on the properties of catanionic vesicular systems: comparison of light-scattering results with ultrasonic and fluorescence spectroscopy HAVLÍKOVÁ, M.; SZABOVÁ, J.; JUGL, A.; MRAVCOVÁ, L.; CHANG, C.; HUANG, P.; PEKAŘ, M.; MRAVEC, F. Colloids and Surfaces A: Physicochemical and Engineering Aspects 202O, vol. 607, December 2020, pp. 1-7 ISSN: 0927-7757 DOI: https://doi.org/10.1016/j.colsurfa.2020.125526 Accepted manuscript © 2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/ dspace.vutbr.cz Study of cholesterol’s effect on the properties of catanionic 1 vesicular systems: comparison of light-scattering results with 2 ultrasonic and fluorescence spectroscopy 3 Martina Havlíková1, Jana Szabová1, Adam Jugl1, Ludmila Mravcová1, Chien-Hsiang Chang2, 4 Po-Sung Huang2, Miloslav Pekař1, and Filip Mravec1,* 5 1 Materials Research Centre, Faculty of Chemistry, Brno University of Technology, Brno, 6 Czech Republic 7 2 Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan 8 * corresponding author,  [email protected] 9 Abstract 10 This work is focused on the study of properties associated with the effect of cholesterol levels on the 11 stability of vesicular systems based on the ion pair amphiphile hexadecyltrimethylammonium12 dodecylsulphate (HTMA-DS) at laboratory temperature. The HTMA-DS catanionic system was doped 13 with dioctadecyldimethylammonium chloride in a 9:1 molar ratio and cholesterol in the amount of 0, 3, 14 13, 23, 33, 43, 53, 63, and 73 mol.% was added. In this system, the size distributions were studied using 15 the dynamic light-scattering technique and the zeta potential was determined. These standard techniques 16 were supplemented by ultrasonic and fluorescence spectroscopy techniques. Due to low stability and 17 high opalescence of samples, spectral techniques were used only for the samples with cholesterol content 18 above 23 mol.%. The results from High-Resolution Ultrasonic Spectroscopy and from Fluorescence 19 Spectroscopy are in agreement. They equally point to a change in the amount of hydration water in the 20 membrane, the largest amount of which is present in the samples with 43 and 53 mol.% cholesterol. 21 Using the light-scattering technique, the short-term stability of prepared vesicular systems was also 22 observed over the first 36 days. Obtained results confirmed that the most stable systems are those 23 containing 43 or 53 mol.% of cholesterol. 24 Keywords: 25 Ion pair amphiphile; Catanionic vesicles; Cholesterol; High-resolution ultrasonic spectroscopy; 26 hydration water; Fluorescence spectroscopy. 27 28 Manuscript Click here to access/download;Manuscript;manuscript Mravec 1.1.docx 1 Introduction 29 Catanionic vesicles have a spherical structure with, hydrophilic and hydrophobic domains. Due to this 30 structure, some hydrophilic, as well as hydrophobic substances, can be encapsulated into these domains 31 [1–6]. The formation of catanionic vesicles is a spontaneous process, but they can be formed also by 32 using semi-spontaneous or mechanical disruption methods [1–5,7,8]. Their structure is very similar to 33 the structure of a liposome, but catanionic vesicles are formed from oppositely charged single chain 34 surfactants or ion pair amphiphile (IPA) [1,3,9,10]. The surfactants usually precipitate in IPA when the 35 oppositely-charged surfactants are mixed in an equimolar ratio [1,6,8]. The vesicles are formed through 36 the interactions between polar headgroups of cationic and anionic surfactants in the aqueous phase and 37 when the counterions are removed from this mixture; the residue is called ion pair amphiphile [1,3,9,10]. 38 Catanionic vesicles may be used as an alternative to liposomes, e.g. such as drug delivery carriers [1– 39 6]. Their advantage is lower price and possibility of preparations from a wide variety of surfactant 40 combinations. However, their size, polydispersity and lamellarity are determined by the choice of 41 surfactants [6]. There also exist so-called catanionic drug-surfactant complexes, which are formed from 42 a surface-active drug and an oppositely charged surfactant, when the drug is a part of a membrane. These 43 vesicles can be an alternative to the encapsulation of drugs into catanionic vesicles [4]. The catanionic 44 drug-surfactant complexes have been intensively studied. For example, among the drug compounds, 45 which have been investigated, there are lidocaine, ibuprofen and naproxen [11] or a combination of 46 diphenhydramine and tetracaine with sodium dodecylsulphate (SDS) as an oppositely charged surfactant 47 [12] for drug release from a gel. The study [13] describes the combination of double-chained surfactants 48 with oppositely charged drug tetracaine hydrochloride. Anionic or cationic surfactants can be combined 49 also with phospholipids, e.g. a combination with dipalmitoylphosphatidylcholine (DPPC) [14], 50 a combination of natural phospholipid soylecithin with IPA which was prepared by mixing SDS and 51 HTMAB [15,16]. Some hybrid vesicles were also prepared from soy phosphatidylcholine and IPA from 52 HTMAB and SDS [15]. 53 However, catanionic vesicles prepared only from IPAs usually have very poor physical stability and 54 they can easily fuse or aggregate together [1,7,9,10] The adjustment of intra/inter-vesicle interactions 55 can help increase vesicles stability. One of the strategies for improving physical stability is the addition 56 of some charged double-chained surfactants. Then, catanionic vesicles are same-charged and the 57 repulsive interactions between them prevail, therefore the aggregation and fusion are inhibited. 58 A different double-chained surfactant then indicates different long-term stability of vesicles [7,9,17,18]. 59 Charged catanionic vesicles can also bring another benefit in enabling the interaction with oppositely 60 charged (bio)polymers through hydrophobic and/or electrostatic interactions. It can also improve higher 61 stability and biocompatibility of them [4,19]. Another step to improve stability is the addition of 62 cholesterol [1,8,9]. Cholesterol is known as a regulator of membrane permeability, elasticity and 63 stiffness and it is able to prevent the aggregation and enhance the stability of bilayers. Cholesterol is the 64 most often-used in liposome as a stabilizing agent. Thanks to the hydroxyl group, it is able to form 65 hydrogen bonds and then it is incorporated into the bilayer of vesicles and increases their stability and 66 fluidity [20,21]. Also the presence of cholesterol can cause the increase in size of particles due to the 67 incorporation to the bilayers and their expansion. However, larger amount of cholesterol may cause the 68 crystallization of carbohydrate chains and the destabilization effect on vesicles [20]. Cholesterol can 69 influence the charge character, the physical stability and the molecular packing of vesicles bilayers 70 [1,8,9,22]. Its inclusion would increase the van der Waals attraction and the average area per IPA’s 71 molecule, and thus reduce the electrostatic repulsion (decrease of zeta potential) [23,24]. The counterion 72 could be gained by the addition of charged double chain surfactants to improve the stability of IPA 73 system. Overall reduction of vesicles zeta potential due to the incorporation of cholesterol leads to the 74 decrease of counterion binding tendency [1,9]. 75 This work discusses the influence of varying cholesterol content on the physical stability and properties 76 of catanionic vesicles. They were composed of IPA and positively charged double chained surfactant 77 dioctadecyldimethylammonium chloride (DODAC). IPA was prepared by mixing 78 hexadecyltrimethylammonium bromide (HTMAB) with oppositely charged sodium dodecylsulphate 79 (SDS). A schematic diagram of the molecular packing of HTMA-DS/DODAC with cholesterol is shown 80 on Figure 1. The average size of vesicles was measured by dynamic light-scattering (DLS) and the zeta 81 potential was measured by electrophoretic light scattering (ELS). These are standard methods that are 82 very often used to describe catanionic vesicles [1,9]. The membrane properties have been studied as 83 well by two methods. One is based on solvent relaxation and the effect on a Laurdan fluorescence probe. 84 Laurdan has been reported many times to be the probe suitable for determining the phase transition of 85 membranes via the interaction of its excited state with polar mobile molecules that penetrate to the 86 vicinity of fluorophore [25]. Due to its hydrophobic tail, the probe is near the polar heads of membrane 87 constituents [26]. The solvent relaxation leads to a bathochromic shift in the fluorescence emission 88 spectra, where the emission maximum in rigid membranes is localized around 440 nm and moves to 89 490 nm when the system becomes fluid [27–29]. The shifts in the Laurdan emission spectrum are 90 quantified by calculating the generalized polarization (GP) value. This value is highest in the case of 91 a rigid system and lowest and also negative in the case of a fluid system [30]. The second method for 92 describing the membrane properties is high-resolution ultrasonic spectroscopy (HRUS). HRUS is an 93 efficient technique suitable for the analysis of colloidal systems [31–34]. It is based on precision 94 measurements of the velocity and the attenuation of high-frequency (ultrasound) sound waves 95 propagating through analysed samples. This technique allows direct non-destructive investigation of the 96 organization of molecules and intermolecular forces in the sample [35]. Also, many properties of 97 surfactants’ activity have been examined using ultrasonic spectroscopy, mainly their interaction with 98 oppositely charged polymers [36–39]. 99 2 Experimental methods 100 2.1 Materials 101 Hexadecyltrimethylammonium bromide (HTMAB) (purity ≥ 98.0%), sodium dodecylsulphate (SDS) 102 (purity ≥ 98.5%) and cholesterol (purity ≥ 99.0%) were purchased from Sigma, USA. 103 Dioctadecyldimethylammonium chloride (DODAC) (purity ~ 97%) was supplied by Alfa Aesar, USA. 104 Chloroform was obtained from Penta s.r.o. A Laurdan fluorescence probe (6-dodecanoyl-N,N-dimethyl105 2-naphthyalamine, purity ≥ 97.0%) was obtained from Sigma, USA and 106 5-Hexadecanoylaminofluorescein (purity ≥ 95.0%) was supplied from ThermoFisher, USA. Pure water 107 was purchased from ELGA LabWater. 108 2.2 Preparation of catanionic vesicles with double-chained surfactant 109 The preparation of IPA has been described in detail in previous literature, e.g. in [1]. Briefly, IPA was 110 prepared by mixing equivalent volumes, 500 mL, of 20 mM aqueous solution of HTMAB and SDS. 111 This mixture was allowed 1 day for complete precipitation. The precipitate was obtained by 112 centrifugation and filtration. The precipitate was further washed on the filter to remove the counterions. 113 After the counterions were removed from this mixture, the residue was denoted as HTMA+DS114 (hexadecytrimethylammonium-dodecylsulphate). The composition was checked and confirmed by the 115 balance of nitrogen and sulphur from Elemental Analysis. 116 To this IPA powder the double-chained cationic surfactant DODAC was added and both were dissolved 117 in chloroform. These components were mixed in the molar ratio of 9:1 and then varying amounts of 118 cholesterol were added. The total amount of cholesterol was set to 0, 3, 13, 23, 33, 43, 53, 63 and 119 73 mol.%. Chloroform was evaporated to form thin films on glass beads. The thin film was then hydrated 120 with 60 mL pure water and this sample was dispersed through the sonification process using an 121 ultrasonic dispersion device (model HD 3 200, Bandelin Electronic GmbH & Co. KG) with an amplitude 122 of 50% and sonification energy of 25 kJ. 123 2.3 Methods 124 Size and zeta-potential. Samples were measured one hour after the preparation without dilution or any 125 other modification. For each of them, the average size of particles and the zeta potential were measured 126 using a ZetaSizer Nano ZS (Malvern Instruments Ltd.) with a 10 mW He-Ne laser source (λ = 633 nm). 127 The measurement was carried out at a constant temperature of 25 °C. The zeta potential was measured 128 by electrophoretic light-scattering (ELS) using a dip cell electrode and calculated by Smoluchowski’s 129 model for the aqueous environment. Each zeta potential measurement was repeated five times. The 130 average size was measured by dynamic light-scattering (DLS) in a glass cuvette. Each average size 131 measurement was performed in triplicate. The value of average size is obtained from the diffusion 132 coefficient D in the Stokes-Einstein equation: 133 𝐷 = 𝑘𝑇 𝑓 ⁄=𝑘𝑇 6𝜋𝜂𝑎 ⁄. (1) Generalized Polarization. For the measurement of generalized polarization it is necessary to add 134 Laurdan such as fluorescence probe and let it solubilize with gentle stirring overnight. The final 135 concentration of Laurdan in samples was kept at 5 x 10–7 M. The samples with high turbidity had to be 136 sonicated in an ultrasonic bath to eliminate the influence of scattering on larger particles. The Laurdan 137 fluorescence emission was monitored with a photon-counting fluorometer FS 5 (Edinburgh Instruments, 138 GB) with temperature controlled by a Peltier module, and the sample homogeneity was maintained by 139 continuous magnetic stirring. A generalized polarization value was calculated from the emission 140 intensities I440 and I490, as follows: 141 𝐺𝑃 =(𝐼440 − 𝐼490) (𝐼440 + 𝐼490)⁄ . (2) High-Resolution Ultrasonic Spectroscopy. Samples were measured one hour after the preparation 142 without dilution or any other modification. The ultrasonic velocity and attenuation for each sample were 143 measured at six selected frequencies in the range from 2.5 to 14.9 MHz using an HR-US 102T ultrasonic 144 spectrometer (Ultrasonic Scientific, Ireland). This device is equipped with two cells enabling single-cell 145 or differential measurements. The differential mode was used in this work. The temperature was 146 controlled with a Haake PC300 heating bath, which provided a temperature stability of ±0.01 °C. The 147 temperature was recorded by a sensor inserted in the ultrasonic spectrometer by the manufacturer. The 148 measuring and the reference cells were filled with deeply degassed sample and water, respectively. 149 Ultrasonic measurements were accompanied by the measurements of density in order to calculate the 150 compressibility from ultrasonic velocity and density. Each value is the average of three replicate 151 measurements and the standard deviation is expressed by error bars. A DSA 5000 M densitometer 152 (Anton Paar, Austria) was used in this work. When combining the ultrasonic velocity (u) with the density 153 measurement (ρ), it is possible to calculate the adiabatic compressibility (β) of the sample using known 154 Laplace equation (3) [36]. 155 𝛽 = 1 𝜌𝑢2 ⁄. (3) Turbidimetry. The turbidity of opalescent samples was taken as the apparent absorbance at wavelength 156 630 nm. It was measured with UV-vis spectrophotometer CARY50 (Varian Inc., USA) and it was 157 calculated similarly as absorbance with the intensity of incident beam I0, and intensity of that light after 158 it passed through the sample I: 159 𝜏 = log⁡(𝐼0/𝐼). (4) 160 pH measurement. The pH of the samples was measured with SevenEasy pH-meter (Mettler Toledo). 161 162 3 Results and Discussion 163 3.1 Characterization of catanionic vesicles 164 3.1.1 Initial average size and zeta potential 165 Most of the samples exhibited significant opalescence, which was problematic for the DLS 166 measurement. Therefore, the turbidimetry was performed using UV-VIS spectroscopy at the wavelength 167 of 630 nm. The turbidity was higher in the sample containing cholesterol from 0 to 23 mol.% and 168 73 mol.% while the lowest value was reported for the samples with 33, 43 and 53 mol. % of cholesterol 169 as shown in Figure 2 and Figure 3. To obtain comparable results, the samples were diluted to the 170 turbidity value between 0.1 and 0.3. The size and the zeta potential were measured by dynamic and 171 electrophoretic light-scattering. The results are listed in Table 1. 172 The samples with 0–23 mol.% of cholesterol showed high instability. This resulted in high standard 173 deviation value so repeated measurements did not lead to comparable results. Only the vesicles with 174 23 mol.% lead to reasonable size results. The aggregation occurs due to system instability, which is in 175 agreement with the turbidimetry measurement. This leads to the formation of polydisperse system, as 176 approved by the polydispersity index value (see Table 1). The generally accepted PDI value for a 177 monodisperse samples is below 0.3 [40]. The instability was confirmed by the measurement of the zeta 178 potential, which is close to instability area. The samples with the cholesterol content of 33 to 53 mol.% 179 had the smallest average size and high zeta potential (see Table 1). With higher cholesterol content (63 180 and 73 mol.%), the stability decreased, probably due to larger amounts of cholesterol which may cause 181 the crystallization of carbohydrate chains and destabilization of vesicles [20]. All samples showed 182 positive zeta potential due to the addition of positively charged double-chained surfactant (DODAC). 183 These results are in good agreement with other published results [1,9,20]. 184 Table 1: Initial average size, polydispersity index, zeta potential and pH of catanionic vesicles HTMA-DS/DODAC 185 with varying molar concentration of cholesterol. 186 Content of cholesterol (mol.%) Initial average size (nm) Polydispersity index (-) Zeta potential (mV) pH (-) 0 572 ± 432 0.474 ± 0.351 44 ± 3 6.3 ± 0.1 3 713 ± 817 0.581 ± 0.195 37 ± 4 6.5 ± 0.1 13 927 ± 473 0.809 ± 0.061 35 ± 2 6.5 ± 0.1 23 272 ± 49 0.496 ± 0.055 38 ± 5 6.5 ± 0.2 33 96 ± 1 0.292 ± 0.049 58 ± 6 6.6 ± 0.2 43 91 ± 1 0.265 ± 0.014 57 ± 7 6.5 ± 0.1 53 98 ± 1 0.265 ± 0.008 56 ± 7 6.6 ± 0.1 63 140 ± 2 0.452 ± 0.080 60 ± 2 6.5 ± 0.1 73 196 ± 12 0.736 ± 0.060 30 ± 2 6.5 ± 0.0 The zeta potential depends on the surrounding conditions such as ionic strength, pH, concentration of 187 additives and temperature [41]. For this reason, the pH of all samples was measured, it was relatively 188 the same in all solutions, there were no significant changes (Table 1). This means that increasing the 189 amount of cholesterol did not affect the pH. Another possibility of influencing the zeta potential is the 190 presence of counterions in the solution [42]. Our vesicles were prepared in pure water so no counterions 191 were present. The only ions in our solutions were chloride anions, they were gained from the dissociation 192 of DODAC. This means that positive charge of vesicles could be influenced (decreased) only by anionic 193 counterions. On charged surface the water molecules are predominantly oriented with respect to the 194 charged surface. This results in a strong local decrease of membrane permittivity [43]. 195 3.1.2 Membrane properties 196 The membrane properties were studied with two approaches. The first one is based on the propagation 197 of ultrasound through a sample and the second one is based on solvent relaxation and the effect on the 198 Laurdan fluorescence probe. 199 The values determined from an ultrasonic spectroscopy are ultrasonic velocity, adiabatic compressibility 200 and attenuation in cell. In a denser (tougher) medium the ultrasound velocity is higher than in less dense 201 medium. In aqueous solutions, the ultrasound velocity reflects mainly the hydration state of solutes – 202 water in hydration shells is less compressible (denser) than bulk water. The reduction in the relative 203 velocity means decreasing difference between the velocity of the solution and solvent. This may mean 204 either higher mobility of hydration water, decrease in rigidity, or increased elasticity of the vesicles 205 membranes. Water has significantly higher compressibility near hydrophobic groups, on the other hand, 206 if it interacts with charged molecules, it has the lowest compressibility [33]. The measured data in 207 Figure 4 show the relative velocity of ultrasonic waves passing through the sample as a function of the 208 cholesterol concentration. As can be seen, the concentration series show a break in the trend of relative 209 velocity around 43 mol.% of cholesterol, where a shallow minimum is formed, indicating the occurrence 210 of the least rigid structures. The density values at 25 °C were also measured for the samples of the second 211 concentration series so that the compressibility can be calculated according to equation (3). 212 The graph in Figure 5 shows that the compressibility values of the system increased with increasing 213 cholesterol content which is interesting because cholesterol should stiffen the membranes. The increase 214 is remarkable particularly in the region where the velocity minimum was observed and up to 63 mol.% 215 where it decreased due to the phase separation in the system. In addition to the values of relative velocity 216 and compressibility, the attenuation parameter of ultrasonic waves (Figure 5) was investigated. The 217 attenuation in the presence of cholesterol is almost constant from 3 up to about 60 mol.% of cholesterol 218 with the value of about 0.5 m-1. At the lowest and highest cholesterol concentration there is a significant 219 increase in attenuation values pointing to increased heterogeneity of samples, which is consistent with 220 their observed opacity or even phase separation. 221 Due to its structure, Laurdan is incorporated into membranes, its fluorophore being located near the 222 polar heads of the membrane components. Its emission spectrum reacts to the presence of hydration 223 water in its surroundings, which is expressed by the quantity of generalized polarization. The lower the 224 GP value, the more hydration water is in the vicinity of the probe [25]. 225 The addition of cholesterol to the vesicles causes an increase in the surface and a decrease in charge 226 density, which will reduce the willingness to interact with countercharged molecules (see Figure 4) [9]. 227 This could mean that there is a decrease in the number of interactions with water (charge-dipole of 228 water), and also the ordered arrangement of water molecules is likely to decrease, due to lower charge 229 density of vesicles [44]. This leads to a decrease in the amount of rigid water, which causes the reduction 230 in the relative velocity. After exceeding a certain concentration of cholesterol in the sample (43 mol.%), 231 this effect is outweighed by the effect of cholesterol on the organization of hydrocarbon chains inside 232 the membrane [9,45]. Due to high opacity of samples with 0, 3, and 13 mol.% of cholesterol, which was 233 also reflected in the measurement of ultrasonic attenuation, because especially in the case of the lowest 234 and the highest cholesterol content in the sample, there was a significant increase in ultrasonic 235 attenuation, indicating increased heterogeneity of the sample. Therefore, the GP values were determined 236 only for the samples with 23–73 mol.% of cholesterol. These values also exhibited a minimum plateau 237 with the cholesterol content of 33 and 43 mol.% related to higher amount of water in the membrane. 238 This plateau is followed by increasing values of GP and decreasing amount of water, which correlates 239 with the results from HRUS. The vesicles with the lowest cholesterol content have the lowest 240 compressibility values, and the membrane compressibility increases with increasing cholesterol content, 241 which is in accordance with the findings of Sarvazyan et al. [33] on charged molecules and their 242 interactions with water dipoles. 243 3.2 Short-term stability of catanionic vesicles 244 The stability of the systems was studied during the first 36 days after the preparation. During this time, 245 the size distribution and the zeta potential were monitored by DLS and ELS measurements. The system 246 instability is manifested by increasing size and polydispersity index values [1,9]. The increase in 247 aggregate size can also be easily monitored using the turbidity measurements. 248 The stability of samples with cholesterol content from 0 to 23 mol.% and 73 mol.% was very poor as 249 was mentioned above. For the sample with 33 mol.%, higher stability was predicted due to high value 250 of the zeta potential, but unfortunately this sample was stable only for one day. On the second day the 251 turbidity and average size increased (see the Supplementary Information), the sample was very 252 polydisperse and the zeta potential decreased. The sample with 63 mol.% of cholesterol was stable for 253 5 days, despite high value of the zeta potential, and after those 5 days the average size of particles was 254 increased by about 30 nm. 255 The samples containing cholesterol with molar concentration of 43 and 53 mol.% were stable over 256 36 days (Figure 6). The average size of particles of those samples was between 80 and 100 nm. The 257 value of zeta potential was between 40 and 60 mV. Some small changes in the average size of particles 258 and zeta potential may have been due to the deviation of measurement. The turbidity of the samples over 259 36 days was checked by UV-VIS spectroscopy (see the Supp. Inf.). The turbidity slightly increased 260 through the time in the case 43 mol. % (Figure S7) sample, but there are no significant changes in 261 53 mol. % sample (Figure S9). 262 4 Conclusions 263 The obtained results show an interesting comparison of methods monitoring the overall parameters of 264 the vesicular system, such as size, zeta potential and methods monitoring outer membrane properties, 265 such as compressibility and the presence of water. One way to influence the overall system stability is 266 the addition of cholesterol. The system with the cholesterol content from 33 to 53 mol.% showed the 267 highest stability. Within short-term stability the most suitable samples appeared to be those with 268 additions of 43 and 53 mol.% In these cases, there were no significant changes in vesicle size, no 269 precipitation was observed and the vesicle system was stable. This was also confirmed by the 270 measurement of zeta potential, which was in stable positive area. The results from high-resolution 271 ultrasonic and fluorescence spectroscopy measurements show that increasing amounts of cholesterol at 272 room temperature do not have a monotonous trend in observed specific quantities. The most stable 273 systems are those where in the outer part of membrane, the highest amount of water is present and 274 exhibits the lowest value of generalized polarization (33, 43 and 53 mol.%). These vesicles membranes 275 are also the most elastic compared to lower or higher contents of cholesterol. 276 As can be seen, this is in a good agreement with short-term stability and confirms the findings from 277 previous work [1,9]. The membrane elasticity, the configuration of water molecules in the solvated part 278 of membrane, and the quantities monitored successfully by high-resolution ultrasonic spectroscopy and 279 fluorescence spectroscopy directly correspond to the expectations about the stability of these vesicles. 280 It is well-known that different amount of cholesterol could influence the phase transition temperature of 281 vesicles and alter their properties e. g. fluidity, stability and leakage of encapsulated compounds 282 [20,46,47]. Below the phase transition temperature, the membrane is in the solid ordered phase and 283 above this temperature, the membrane is in the liquid disordered phase [48]. Thanks to the inclusion of 284 cholesterol, new phase is formed – liquid ordered [49]. Better understanding of this phase behaviour 285 could help design the most suitable composition of IPA’s vesicles. These are the reasons why we plan 286 to extend these experiments with more detailed description of behaviour at different temperatures. 287 Acknowledgement 288 This work was supported by the Czech Science Foundation, project No. 19-14024J (GACR), and 289 Ministry of Science and Technology, Taiwan, project No. MOST108-2923-E-006-006-MY3. 290 291 mol.%) Figure 3 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 020 40 60 80 Relative velocity (m/s) Generalized polarization (-) Cholesterol content (mol.%) Generalized polarization Relative velocity Figure 4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 4.474E-10 4.475E-10 4.476E-10 4.477E-10 4.478E-10 4.479E-10 4.480E-10 4.481E-10 4.482E-10 4.483E-10 020 40 60 80 Attenuation in cell (1/m) β(Pa-1) Cholesterol content (mol.%) Compressibility Attenuation in cell Figure 5 35 45 55 65 75 85 95 0 40 80 120 160 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 Zeta potential (mV) Average size (nm) Time (days) 43 mol.% Average size 53 mol.% Average size 43 mol.% Zeta potential 53 mol.% Zeta potential Figure 6 Figure 1: A schematic diagram showing the influence of cholesterol on the molecular packing of HTMA-DS/DODAC. Adapted from [9]. Figure 2: The dependence of turbidity of vesicular system on the cholesterol content 1 hour after the sample preparation. Figure 3: The picture of samples with different amount of cholesterol (from 0 to 73 mol. %). The lowest turbidity was observed for the sample with 33, 43 and 53 mol.% of cholesterol. Figure 4: The dependence of generalized polarization and ultrasound relative velocity on the cholesterol content. Figure 5: The dependence of compressibility and attenuation in cell on the content of cholesterol. Figure 6: The dependence of average size and zeta potential of particles on the time of measurement for the sample containing 43 and 53 mol. % mol. of cholesterol. Figure Titles Supplementary Material Click here to access/download Supplementary Material Supplementary information.docx CRediT author statement Martina Havlíková: Investigation, Writing - Original Draft. Jana Szabová: Investigation. Adam Jugl: Investigation, Formal analysis. Ludmila Mravcová: Conceptualization, Formal analysis, Project administration. Chien-Hsiang Chang: Conceptualization, Writing - Original Draft. PoSung Huang: Conceptualization. Miloslav Pekař: Conceptualization. Filip Mravec: Conceptualization, Supervision, Funding acquisition, Writing - Review & Editing. Credit Author Statement Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: *Declaration of Interest Statement