1 Introduction

In the basis of many specific processes in biological systems, the intermolecular interactions are found. Relatively simple, synthetically available molecules which are clever in reversible switching of different protein functions can help in fundamental understanding of processes in living systems. They can also work as perspective therapeutic agents to correct some pathological statuses connected with “failures” in biomacromolecule functioning [1]. Significant progress has been made in the development of small molecules which can assist in transport of ions, phospholipids and proteins across membrane barriers [2]. The design of synthetic molecules which can bind to a protein surface and block biologically important protein–protein interactions remains a major challenge, too [3]. The development of synthetic receptors for selective binding of biomolecules, which are not able to permeate the cell membrane without assistance, is the first step to create transfection agents [4].

One of the most perspective bases for engineering of synthetic receptors, modifying and binding agents are calixarenes, offering such advantages as the presence of multivalent domains, the conformational lability and the possibility for functional adaptation. Calixarenes are cup-shaped molecules that can form inclusion complexes with a wide range of guest species. Each calixarene contains repeating phenolic units organized into macrocycle via methylene bridges [5, 6]. The calixarene “cups” have a vase-like structure formed by an upper and lower rims and central annulus. The polar and nonpolar features of their cavities enable calixarenes to interact with a wide range of guest species, depending on the binding groups substituted at each rim and the number of repeating units in the macrocycle. So, they belong to a broad range of compounds, such as crowns, cryptands, and cyclodextrins, which are already well known to form host–guest complexes in solution [7] and to display biological activity of different classes of antibiotics concerning their interaction with lipid membranes [8]. One should especially note the conformational variety of calixarenes such as cone, partial cone, 1,2- and 1,3-alternates.

The replacing of methylene bridges between the aromatic units in calix[4]arenes by sulfur atoms leads to thiacalix[4]arenes [9–11]. The presence of four sulfur atoms results in many new features as compared to “classical” calixarenes, including different coordination behavior, facile oxidation of bridge sulfur atoms, different size and different conformational behavior. Thiacalix[4]arenes exhibit a broad range of interesting functions, which makes this compound fundamentally interesting and promising for a variety of applications [10–12]. Thiacalixarene chemistry dates back to 1997 with the first report on p-tert-butylthiacalix[4]arene, achieved through a stepwise procedure in 4% yield by Sone et al. [13]. In the same year, Miyano and co-workers [14] reported a facile procedure which involved heating a mixture of p-tert-butylphenol, elemental sulfur (S8) and NaOH in tetraethylene glycol dimethyl ether to afford p-tert-butylthiacalix[4]arene. Thiacalixarenes [11, 12], sulfur-bridged cyclic oligo-phenols, have been one of the most studied supramolecules in recent times. However, contrary to calixarenes, thiacalixarenes are not so trivial to synthesize and apart from a few direct synthetic methods [14–19] little has been reported concerning their synthesis, thus their chemistry is still an enigma. This is one of the reasons that properties of thiacalixarenes are studied not broadly.

Thiacalix[n]arenes provide a unique opportunity for modification of macrocyclic platform and for wide variation of hydrophilic-lipophilic properties. This circumstance is especially important because the selective modification at the upper rim of a macrocyclic platform by suitable heteroatom’s groups often increases the efficiency and selectivity of binding. Among synthetically available stereoisomers of thiacalixarene, the 1,3-alternate which contains two substitutes at different sides of the macrocycle, is of the interest for the receptor engineering.

In the present work, we studied the influence of the structure of functional groups of tetrasubstituted at the lower rim p-tert-butyl thiacalix[4]arenes in the 1,3-alternate configuration on its interaction with model lipid membrane. For this purpose, two complementary physical methods, electron spin resonance (ESR) spin-label technique and Fourier transform infrared spectroscopy (FTIR), were applied to study the alterations in the structure of a bilayer at different concentrations of calixarenes 1–3 (Fig. 1). Two additional techniques, 1H nuclear magnetic resonance (NMR) and transmission electron microscopy (TEM), were applied for confirmation of calixarene and liposome structure, respectively.

Fig. 1
figure 1

Chemical structure of studied calixarenes

Full size image

2 Experimental

2.1 Chemicals

1,2-dipalmitoyl phosphatidylcholine (DPPC), molecular weight (M.w.) of 734.1, reagent grade (99%) from Sigma (P0763) was used as received. 5- and 16-doxylstearic acid spin probes, named hereinafter as 5-DSA and 16-DSA, were purchased from Sigma. Chloroform (ClF) of analytical grade was from ACROS Organics.

The following thiacalix[4]arenes (Fig. 1) were synthesized according to the procedure described in Ref. [20]:

  1. 1.

    5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetra[(N-propylmorpholine)amidocarbonyl)-methoxy]-2,8,14,20-tetrathiacalix[4]arene in 1,3-alternate configuration. M.w. 1458.01, melting point (Mp) 150°C. 1H NMR (CDCl3, δ, ppm, J/Hz): 1.21 (36H, s, (CH3)3C), 1.72–1.80, 2.32–2.53, 3.24–3.38, 3.57–3.72 (56H, m, –HN(CH2)3N, –N(CH2CH2)2O), 3.99 (8H, s, OCH2O), 7.52 (8H, s, ArH), 8.14 (4H, t, 3 J HH = 5.1 Hz, NH).

  2. 2.

    5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetrakis[3-N,N-bis(3-aminopropyl)-aminopropoxy]-2,8,14,20-tetrathiacalix[4]arene in 1,3-alternate configuration. M.w. 1406.2, Mp 137–139°C. 1H NMR (CDCl3, δ, ppm, J/Hz): 1.23 (36H, s, (CH3)3C), 1.38 (16H, s, NH2), 1.59 (12H, m, CH2–CH2–CH2), 1.80 (12H, m, CH2–CH2–CH2), 2.46 (24H, t, 3 J HH = 7.1 Hz, N–CH2), 2.73 (16H, t, 3 J HH = 6.9 Hz, CH2–N), 4.00 (8H, t, 3 J HH = 7.9 Hz, O–CH2), 7.40 (8H, s, Ar1-H).

  3. 3.

    5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetrakis[3-N,N-di(hydroxycarbonyl) aminopropoxy]-2,8,14,20-tetrathiacalix[4]arene in 1,3-alternate configuration. M.w. 1413.73, Mp 215–217°C. 1H NMR ((CD3)2SO, δ, ppm, J/Hz): 1.20 (36H, s, (CH3)3C), 1.33 (8H, tt, CH2–CH2–CH2), 2.64 (8H, t, 3 J HH = 5.7 Hz, N–CH2), 3.38 (16H, s, O–CH2–CO), 3.86 (16H, t, 3 J HH = 6.3 Hz, O–CH2), 7.34 (8H, s, Ar–H).

2.2 Sample Preparation

Unilamellar liposomes were prepared as described elsewhere [21]. For calixarene-free samples, 12 or 26 mg of DPPC were placed into the flat-bottomed glass vessel, dissolved in 200 μl of ClF and then dried overnight with a vacuum evaporator. The obtained film was dissolved in 500 μl of 20 mM Tris–HCl buffer (pH 7.5) containing 150 mM of NaCl with the following stirring to an isotropic suspension. The suspensions were subjected to 5 cycles of freezing in liquid nitrogen and thawing at 55°C, followed by 19 times extrusion through polycarbonate filter with 100 nm pore size (Avanti 610005) at 55°C. The obtained unilamellar liposomes had the sizes in the range of 80–100 nm, according to the TEM data (Fig. 2).

Fig. 2
figure 2

TEM micrographs of liposomes: a DPPC, b DPPC + 1, c DPPC + 2, d DPPC + 3. DPPC concentration is 70 mM, drug-to-lipid molar ratio is 1/70. Bar 100 nm

Full size image

Liposomes containing calixarene were prepared according to the same protocol with the following distinctions. The hydrophobic calixarene 1 was dissolved in ClF together with DPPC, dried and then dissolved in the buffer. Calixarenes containing charged groups were preliminarily dissolved in low volume of water which was acidified to pH 2 for sample 2 or alkalized to pH 12 for 3 and then adjusted to a wanted concentration by buffer. The following molar ratios of calixarene to lipid were used:

  • [1]/[DPPC] = 1/500, 1/50;

  • [2]/[DPPC] = 1/500, 1/50;

  • [3]/[DPPC] = 1/50.

Ethanol solutions of spin probes were dried at room temperature. Prepared dispersions of liposomes were added to the dry film of spin probe and shaken at 55°C up to full dilution of sediment. In all samples, the spin probe concentration was equal to 10−4 M. In ESR experiments, the concentration of DPPC was equal to 20 mM. For FTIR measurements, the concentration of lipid was equal to 70 mM and the samples were free of spin probes.

2.3 NMR Measurements

Bruker AVANCE 400 NMR spectrometer was used to prove chemical structure of synthesized calixarenes.

2.4 TEM

The prepared liposome suspensions were processed using copper grids to adsorb liposome particles from the suspension, then stained in 2.5% uranyl acetate for 30 s and dried [22]. The specimen was observed under JEM 1200EX TEM microscope (JEOL, Japan) operated at 80 kV.

2.5 ESR Measurements

The ESR spectra were recorded on a RE1306 radiospectrometer (USSR) with the accuracy of temperature control of ± 0.5°C. All spectral measurements were performed starting from high temperatures. ESR spectra were registered under conditions of slow passage through resonance and in the absence of saturation. In this case, the high-frequency modulation amplitude did not exceed 0.2 G and the signal-to-noise ratio was no more than 15. The error in determination of the distance between the peaks in the ESR spectra depended on their width, the highest possible error was ± 0.3%. The details of spectral analysis were described by us previously [23, 24]. To characterize a dynamic structure of the systems studied we used the maximum hyperfine (hf) splitting of 2A max in the ESR spectra. The isotropic hf splitting constant, a N, was used to estimate the polarity of the nitroxide moiety environment. The a N values were obtained from the ESR spectra as follows [25]. When the spectrum consists of three lines, the a N value is derived directly from the spectrum by standard procedure. When the spectrum contains the outer wide peaks (OWP) it is possible to estimate a N as:

$$ a_{\text{N}} = 1 / 3 { }(A_{||} + 2A_{ \bot } ), $$

where 2A ||  = 2A max is the maximum hf splitting in the ESR spectrum, while 2A ⊥ ≈ 2A min (Fig. 3).

Fig. 3
figure 3

ESR spectra of 5-DSA for dispersions of liposomes prepared from DPPC (a), DPPC + 1 (b) and DPPC + 2 (c). T = 15°C

Full size image

2.6 FTIR Spectroscopy

The IR spectra were recorded on a Tensor-27 FTIR spectrophotometer (Bruker) with 4-cm−1 resolution and 128 scans accumulation. The samples were placed in a temperature-controlled demountable CaF2 cell with 10-μm space gap. Temperature was varied in the range of 15–55°C with an accuracy of ±0.05°C. Samples were kept at each temperature for 7 min before spectra recording. The water vapor and solvent absorptions were subtracted from the experimental spectra. No smoothing was implemented. Spectra were processed using the spectrometer software.

Thermodynamic parameters of lipid phase transition were extracted from the experimental temperature dependences of the stretching vibration frequencies of the lipid methylene groups fitted by the equation

$$ y = \left( {\left( {a \times T + b} \right) + \left( {c \times T + d} \right) \times \exp \left( {\left( {\Updelta H/RT} \right) \times \left( {\left( {T - T_{\text{m}} } \right)/T_{\text{m}} } \right)} \right)} \right)/\left( {1 + \exp \left( {\left( {\Updelta H/RT} \right) \times \left( {\left( {T - T_{\text{m}} } \right)/T_{\text{m}} } \right)} \right)} \right), $$
(1)

where T is the operating temperature in degrees centigrade, a and c are the adjustable parameters which characterize the slope of the observed dependence below and above the transition region, respectively, b and d mean the frequency values at T = 0, R is the gas constant, T m is the half-transition temperature and ΔH is the apparent Vant-Hoff enthalpy.

3 Results and Discussion

3.1 TEM

The vesicles made from DPPC were firstly characterized by TEM (Fig. 2). The sizes of prepared liposomal particles vary in the range of 70–100 nm and most of them look unilamellar. No appreciable changes in vesicles, neither aggregation nor visible fusion, were detected in the presence of calixarenes 1–3.

3.2 ESR Spectroscopy

The liposome membranes which are studied in the present work constitute DPPC bilayers. By analogy with other amphiphilic surface-active spin probes [26], it is reasonable to assume that 5-DSA and 16-DSA are incorporated in the lipid bilayer with their polar head groups being anchored to the lipid-water interface. At that, the nitroxide moiety of spin probes in the 5th position is disposed closer to the polar surface of the bilayer and in the 16th position it is located in central, hydrophobic domain of the bilayer.

Figures 3 and 4 depict, respectively, the ESR spectra of 5-DSA and 16-DSA in DPPC-based water suspensions of liposomes in the absence of calixarene and in the presence of compounds 1 and 2. The ESR spectra for systems DPPC/3, which are not shown here, have the similar shape. The common feature of the ESR data for all systems studied is the presence of outer wide peaks in the spectra due to the low mobility of nitroxide moieties of spin probes. The maximum hf splitting, 2A max, determined from the analysis of such spectra can be used to characterize the dynamical structure of a lipid bilayer. It is well known [26] that 2A max is related linearly with an order parameter S. The S values characterize a degree of space ordering of nitroxide z-axis and, as a consequence, are sensitive both to the amplitude and to the rate of the nitroxide moiety motion [27, 28]. Since the z-axis of the N–O group in the fatty acid spin probe is parallel to the long axis of a probe molecule, the order parameter S (and consequently the 2A max value) characterizes the ordering of lipid molecules in the bilayer. The higher is the 2A max value, the more tightly lipid molecules are packed in the bilayer, and the smaller is the mobility of nitroxide moiety of spin probes.

Fig. 4
figure 4

ESR spectra of 16-DSA for dispersions of liposomes prepared from DPPC (a), DPPC + 1 (b) and DPPC + 2 (c). T = 15°C

Full size image

The values of 2A max for different nitroxide moiety positions in spin probes are shown in Fig. 5. The data show that at 35°C (below the gel-to-liquid crystalline transition in DPPC at ~42°C) the packing density of lipids in liposomes containing calixarenes exhibits changes. The most appreciable influence is apparent in the vicinity of the polar domain (5-DSA) and the least is detected in the hydrophobic zone (16-DSA) of the bilayer. The compound 3 slightly increases the degree of lipid ordering (packing), whereas the compounds 1 and 2 decrease their ordering. The disordering effect of 1 exceeds that of 2.

Fig. 5
figure 5

Maximum hf splitting, 2A max, in the ESR spectra for spin probes 5-DSA and 16-DSA in liposome dispersions without calixarene (control) and in the presence of compounds 1, 2 and 3. T = 35°C

Full size image

The value of the isotropic hf splitting constant a N could be used to estimate the polarity of the spin probe microenvironment: the higher is the polarity, the higher is the a N value. For example, for nitroxide probe in water and heptane a N values are equal to 16 and 14 Hz, respectively [29]. The comparison of these values with the data presented in Fig. 6 shows that the environment of a spin probe in the bilayer formed from pure DPPC is noticeably hydrophobic, and under the transfer from 5-DSA to 16-DSA the hydrophobicity approaches the value of heptane. This proves the assumption about elongated conformation of hydrocarbon chains of the spin probes used and the location of nitroxide moiety of 16-DSA in the center of the bilayer. The introduction of calixarenes into the system influences the a N value. These alterations appear to be more pronounced in the vicinity of the bilayer polar interface rather than in its central domain. It agrees with the fact that the local polarity may be influenced either by penetration of water or exogenic polar molecules. Close examination of data in Fig. 6 shows that maximal increase in a N is observed for compound 3 which contains acid groups and the minimum effect is found for compound 1 which is the most hydrophobic among 1–3.

Fig. 6
figure 6

Isotropic hf splitting constant, a N, in the ESR spectra for spin probes 5-DSA and 16-DSA in liposome dispersions without calixarene (control) and in the presence of compounds 1, 2 and 3. T = 35°C

Full size image

3.3 FTIR Spectroscopy

Figures 7, 8 and 9 show temperature dependences of spectral parameters of functional groups of lipid molecules which can be used to characterize the state of different bilayer regions. The frequency of methylene stretching vibration is related to the extent of trans-gauche isomerization in methylene chains [30] and is widely used to probe ordering in hydrophobic region of the lipid bilayer. To monitor the lipid phase transitions by FTIR spectroscopy, the temperature dependence of ν s (CH2) is commonly used since this band is sufficiently intense and free from superposition of other vibration bands. As follows from Fig. 7, in the vicinity of gel-to-liquid crystalline phase transition (~42°C) the frequency of methylene groups absorption increases sharply in consequence of a drastic increase in mobility of the lipid fatty-acid tails under gel melting. The thermodynamic parameters of pure DPPC bilayer transition determined from Eq. (1) (T m = 42.07 ± 0.04°C; ΔH = 36.9 ± 3.6 kJ M−1) are very close to the published data of 41.3 ± 1.8°C and 34.3 ± 5.9 kJ M−1 [31]. The results presented in Fig. 10 show that studied calixarenes influence the transition temperature in different extent. The observed changes are specific for each of the added compounds and display the dose-dependent behavior relative to the calixarene-to-lipid molar ratio. The most hydrophobic compound 1 lowers noticeably the T m value, the strongly basic calixarene 2 practically does not influence this parameter and the compound 3 containing acidic groups slightly increases the transition temperature. Though the changes in T m value are relatively small they still exceed the data error.

Fig. 7
figure 7

Temperature dependence of the symmetric stretching vibration frequency of methylene groups, ν s (CH2), of the fatty-acid tail radicals of DPPC in liposome dispersions without calixarene (control) and in the presence of compounds 1, 2 and 3 in the molar calixarene-to-lipid ratio of 1:50. Lines show the results of fitting of experimental data by Eq. (1)

Full size image
Fig. 8
figure 8

Temperature dependence of the stretching vibration frequency of carbonyl group, ν(C=O), of DPPC in liposome dispersions without calixarene (control) and in the presence of compounds 1, 2 and 3 in the molar calixarene-to-lipid ratio of 1:50. Lines show the results of fitting of experimental data by Eq. (1)

Full size image
Fig. 9
figure 9

Temperature dependence of the symmetric stretching vibration of phosphorus residues, ν s (PO2−), of the lipid head groups of DPPC in liposome dispersions without calixarene (control) and in the presence of compounds 1, 2 and 3 in the molar calixarene-to-lipid ratio of 1:50

Full size image
Fig. 10
figure 10

Dependence of the transition temperature, T m, on the molar calixarene-to-DPPC ratio according to the frequency dependence of the stretching vibration of methylene groups ν s (CH2) of the fatty-acid tail radicals of lipids

Full size image

The effect of calixarenes on the interfacial and polar head group regions of the DPPC bilayer has been elucidated from the temperature dependences of carbonyl, ν(C=O), and phosphate, ν s (PO2−), stretching frequencies (Figs. 8 and 9). The hydration is the main factor affecting the frequency of the absorption band of lipid polar groups. It is well established that considerable and sharp decrease of frequency at the transition is caused by increase of hydration of corresponding groups under gel melting [30]. The difference in the band position between the systems studied beyond the transition region may also be explained by different hydration induced by alterations in conformation of lipid head groups due to the drug influence. Below the transition (Fig. 8), an increase in the absorption frequency of ester carboxyl groups is observed for all studied systems, suggesting partial dehydration of carboxyls. The most significant frequency shift is caused by drug 1 which is most hydrophobic among the studied calixarenes, but at the same time carries the hydrogen bond donating groups at the upper and lower rims. Note that it is just the same drug which lowers the transition temperature (Fig. 10). At temperatures above the transition, the differences between curves become more ambiguous and we withhold comments.

To some extent, a different picture is observed in the region of DPPC phosphate group (Fig. 9). The acidic compound 3 does not influence the spectrum of phosphate residue at all. On the contrary, basic compounds 1 and 2 decrease the frequency of this band to nearly equal extent suggesting direct binding to this group.

4 Conclusions

Summarizing the data presented above it is possible to suppose the following interaction model for calixarenes with a lipid bilayer. The binding of calixarenes takes place predominantly in the polar domain of the bilayer. The compounds 1 and 2 interact with phosphate groups. Compound 1, being the most hydrophobic among the three calixarenes studied, partly or completely penetrates into the bilayer, promoting the bilayer disorder and decreasing the transition temperature. More hydrophilic substance 2 settles oneself close to the surface of the bilayer and does not influence visibly the transition temperature. We can speculate that calixarene 3 forms bridges between positively charged choline groups of adjacent lipid molecules on the surface of a bilayer. In this case, the appearance of additional ties between lipid heads favors to their more dense packing and results in increase of the phase transition temperature.