Файл: 0002W.pdf

1. INTRODUCTION

A characteristic feature of many chemicals of physiological importance is their diblock structure arising from the presence of two segments of a very different chemical characterpolar and apolar - within one molecule. Such a structure of molecules, called amphiphilic or amphipathic, results in their spontaneous accumulation and self-organization at interfaces (accompanied by changes in surface/interfacial properties), a phenomenon which is called surface activity [1-3]. Among various interfaces, one of the most abundant in nature is the air/water interface.

The behavior of surface active molecules (surfactants) at the air/water interface depends on the length of their hydrocarbon part and the structure and chemical character of their polar moiety (charged or not). For example, short-chained amphiphiles (usually those possessing less than 12 carbon atoms in the hydrophobic part) are water soluble. Upon dissolving in water, they undergo adsorption from the bulk into the interface and gather at the phase boundary, forming so-called soluble (adsorbed; Gibbs) monolayers (films). However, when the hydrocarbon chain lengths of molecules increase, their solubility in water is drastically decreased. However, when such molecules are dissolved in non-aqueous, organic solvent, immiscible in water and dropped onto the water surface, they form - after solvent evaporation - a monomolecular layer called Langmuir (insoluble or floating) monolayer (film) [4].

The phase boundary is thus responsible for an unequal distribution of surface active molecules and also determines their particular orientation [5], due to different properties of phases that are in contact. Such a natural barrier separating the inside of each living cell from the outside is cellular membrane, which - apart from many other functions in cellular processes (like transport, communication and regulation) - allows cells to selectively interact with their environment and provides a site of action for a number of biomolecules, e.g. hormones, enzymes or drugs [6, 7]. Surface activity of a drug facilitates its affinity to the cellular membrane.

Studies on molecular interactions are of utmost importance in searching for the mode of action of drugs acting at membrane level, their selectivity and toxicity [8]. The drug-membrane interactions determine orientation, conformation and partitioning of the drugs in

*Address correspondence to this author at the Jagiellonian University, Faculty of Chemistry, Ingardena 3, 30-060 Kraków, Poland; Tel: +48-126632082; Fax: +48-126340515; E-mail: ucdynaro@cyf-kr.edu.pl

membranes and, in consequence, have a high impact on such vital processes as transport, distribution, accumulation and efficiency of the drug. These interactions can also modify the physicochemical properties of biomembranes, thus influencing their functioning. The latter may be directly related to the mechanism of action of the drug or its toxicity.

The interactions between chemicals and membranes can be studied on natural systems, either isolated or not (living cells), or on membrane models. The latter have advantage of being simple and well-defined, thus enabling to study a specific aspect of a given phenomenon, contrary to highly variable and quite complicated natural systems, which provide only general information on the particular problem of interest [9].

Many different membrane models (reviewed in refs. [10-12]) have been applied to investigate the interactions between biochemicals and membrane components. Most popular are: Langmuir monolayers [9, 10], liposomes (vesicles) [13], black lipid membranes (BLM) [14] or surface - confined membrane systems [15], including solid-supported lipid membranes, hybrid bilayers or polymer-cushioned lipid bilayers. In this paper we report on a successful application of the Langmuir monolayer technique, which was indicated in many papers as a potent method for mimicking of cellular membranes [9, 10, 16], to study the interactions between edelfosine - a new generation anticancer drug - and membrane lipids, aimed at getting insight into the drug’s mode of action and its selectivity. This method has successfully been applied to investigate the behavior of various membrane-active drugs [see e.g. 17-20], including other new generation anti-cancer lipid (like for example miltefosine [21, 22]) in model lipids systems.

The detailed description of the Langmuir monolayer technique can be found elsewhere [4, 23]. Briefly, the technique of monomolecular layers formed at the aqueous solution-air interface, termed the Langmuir monolayer technique, is a method in which an amphiphilic molecule of a proper hydrophilic-hydrophobic balance (resulting from the presence of a polar “head”, anchoring molecule at the water surface, and an apolar “tail” long enough to prevent the dissolution of the molecule in bulk water) is placed onto a water surface to form a monomolecular layer, called Langmuir monolayer/film (Scheme 1). In a common experiment, a known amount of surface active compound dissolved in water-immiscible, volatile organic solvent (such as chloroform) is dropped onto the water surface. After evaporation of the solvent, the molecules cover the entire surface forming a monolayer, which is then compressed

with movable barrier(s), using the Langmuir trough. In a classical experiment this process is usually monitored with the surface pressure (π) - area (A) isotherm.

The application of the Langmuir technique in biomedical sciences results from the fact that the cellular membrane components as well as the majority of biomolecules posses an amphiphilic structure, enabling their surface activity, and are capable of Langmuir monolayer formation. Therefore, multicomponent Langmuir films formed at the air/water interface by membrane components may serve as a simple model of biomembrane. Of course, Langmuir monolayer represents only one leaflet of a bilayer membrane, however, strong correlation has been found between monolayers and bilayers prepared from cellular membrane components. It was found that lipid monolayers reveal similar properties (such as pressure, area per lipid molecule, phase transition, elastic compressibility) to bilayers at surface pressures of 30-35 mN/m (so-called monolayer-bilayer correspondence) [16, 24]. Therefore, the results of Langmuir monolayer experiments at these conditions can be successfully linked to a bilayer system. By mixing the Langmuir monolayer components in a certain proportion it is possible to prepare either various types of membranes or individual membrane leaflets, while maintaining their individuality and asymmetry. Additionally, by mixing membrane lipids and

water-insoluble, amphiphilic drugs (or other biomolecules) in a Langmuir monolayer, it is possible to investigate the effects of these chemicals on membrane organization. This can be done by measuring the surface pressure-area (π-A) isotherms during compression of the mixed films containing components in various proportion. However, such a basic characteristic is usually not sufficient and must be complemented by analyzing the miscibility and intermolecular interactions of the investigated mixed system(s). Plotting such parameters as the mean molecular area (A12) or the excess area (A12Exc) versus the monolayer composition (represented usually as mole fraction of component 1(2), X1(2)) is a qualitative way to examine the interactions between molecules, i.e. we can identify weather the interactions are repulsive, attractive or the system behaves ideally. By plotting the collapse pressure values (πcoll) as a function of film composition we can easily verify whether the investigated molecules mix in monolayer or not. Values of the compression (elastic) modulus (Cs-1) provide information on the physical state (e.g. gaseous - G, liquid - L, solid - S, etc) and ordering of the monolayer. Moreover, by plotting this parameter as a function of the surface pressure it is possible to identify the existence of phase transitions between the respective states (e.g. gaseous (G) - toliquid expanded (LE), liquid-expanded (LE)-to-liquid condensed (LC)), which manifest as minima [25].

Edelfosine in Membrane Environment - the Langmuir Monolayer Studies

The compression modulus values calculated for mixed films compoted of lipid and drug molecules allow analyzing the influence of the latter component on the lipid monolayer’s fluidity. A quantitative way to examine the interactions between monolayer components is the calculation of thermodynamic parameters, such as for example excess free enthalpy of mixing, GExc. Different approaches used to characterize molecular interactions in monolayers are reviewed in more details in Ref. [26], while the paper by Gong et al., [27] can serve as a representative paper, providing very useful experimental examples of the analysis of intermolecular interactions using the Langmuir technique. Additional valuable information on the properties of the investigated monolayer can be obtained when the Langmuir technique is combined with other experimental methods, e.g. Brewster angle microscopy (BAM) [28, 29], atomic force microscopy (AFM) [30, 31], polarization modulation infrared reflection absorption spectroscopy (PM-IRRAS) [32, 33], the pendant drop technique to measure dynamic elasticity [34, 35] or the grazing incidence X-ray diffraction (GIXD) method [36, 37]. For a successful application of the Langmuir monolayer technique that helped to get insight into mode of action of bioactive molecules, see for example Refs. [38-42].

In this paper we review the results of the Langmuir monolayer experiments performed for edelfosine (ED):1-O-octadecyl-2-O- methyl-rac-glycero-3-phosphocholine, in short Et-18-OCH3 - a new generation anticancer drug belonging to alkyl-lysophospholipids (ALPs), which are metabolically stable, synthetic ether-linked analogs of lysophosphatidylcholine [43]. Contrary to traditional cytotoxic drugs applied in chemotherapy, ALPs do not target DNA but cell membrane since - due to structural similarity with natural phospholipids - they can adsorb and accumulate in biomembrane, which is their primary site of action [44]. Due to its amphiphilic structure, edelfosine is surface active and can be dissolved in water, forming adsorbed (Gibbs) monolayers [45] but, interestingly, is also capable of Langmuir monolayer formation [46]. Since basic membrane components (like phospholipids [47] and sterols [48]) are also surface active and possess film-forming properties, therefore the Langmuir monolayer technique can well be applied in investigating behavior of edelfosine in membrane lipids environment. Although many articles concerning antitumor effectiveness of edelfosine have appeared in recent years, many questions regarding this drug still remain open. For example, in vivo or in vitro studies on cell lines have not clarified so far which of membrane components may target the drug to the neoplastic cell and is responsible for such a high selectivity of this drug. Why edelfosine is active on some kinds of cells while its effectiveness on another lines is significantly reduced? In this paper we show that the interactions between edelfosine and membrane components, successfully examined with the Langmuir monolayer technique, helped to answer these questions.

2. EDELFOSINE IN MONOLAYERS

The monolayer experiments performed so far on edelfosine involved the characteristics of its pure films, interactions of the drug with particular membrane lipids as well as their mixtures. Below the results of the foregoing experiments are presented and analyzed from the point of view of edelfosine mechanism of action and its selectivity.

2.1. Surface Active Properties of Edelfosine

Surface activity of edelfosine was found to be comparable to other single-chained phospholipids (e.g. lysophosphatidylcholine). As already mentioned above, edelfosine was found to form both adsorbed (Gibbs) monolayers [45] as well as insoluble (Langmuir) films [46].

Edelfosine dissolves in water and behaves as typical soluble surfactant, i.e. it adsorbs at the air/water interface and - with increasing concentration - decreases surface tension of water,

Anti-Cancer Agents in Medicinal Chemistry, 2014, Vol. 14, No. 4 501

finally forming micelles at the critical micelle concentration (CMC) of 5-10 µM as detected with the Langmuir balance [45]. Edelfosine can also penetrate from water into insoluble monolayer or bilayer from phospholipids (POPC) [45]. Such an incorporation of edelfosine causes solubilization of phospholipid molecules from monolayers/bilayers and provokes the formation of mixed edelfosine-phospholipid micelles. This is referred to the so-called detergent activity, which is rather low for edelfosine [45].

Apart from adsorbed monolayer formation, edelfosine can also form insoluble films of high stability when spread onto the air/water interface [46]. In a comprehensive study [46], the properties of edelfosine in Langmuir monolayers at various experimental conditions were investigated, in addition to its molecular organization, film thickness and electrical properties. The state of edelfosine film was classified as liquid expanded (LE). The visualization of edelfosine film with Brewster angle microscope (BAM) proved homogenous distribution of molecules at the surface and only at higher surface pressures (close to the collapse point) small domains were observed. It was also found that near the monolayer collapse edelfosine molecules orient vertically in respect to the surface - the thickness of edelfosine film in this pressure region correlates with the length of hydrophobic part of the drug molecule. Excellent film-forming properties of edelfosine, especially high stability of its monolayers, make the Langmuir monolayer technique a suitable tool for its studies.

2.2. Edelfosine in Lipid Monolayers

Most frequently studied mixed monolayer systems consist of two components (binary films) and despite compositional complexity of biological membranes, they are useful to characterize the interactions of surface active biomolecules with individual membrane lipids. Although the respective membrane leaflet can be better modeled with multicomponent monolayers (3 or 4 component systems), such experiments are not frequent, due to complexity in analyzing of the obtained results. However, edelfosine has been investigated in mixtures with various membrane lipids both in binary systems as well as in ternary and quaternary monolayers. The obtained results are summarized below.

2.2.1. Binary Mixed Monolayers of Edelfosine/Membrane Lipid

Natural membranes are composed of numerous lipids, which can be classified as: phosphoglycerides (phosphatidylcholines, PCs; phosphatidylethanolamines, PEs; phosphatidylserines, PSs; phosphatidylinositoles, PI), sphingolipids (sphingomyelin, glycolipids), and sterols [6]. The concentration of the respective lipids in biological membrane is varied (for example gangliosides, which belong to the group of glycolipids - in comparison to PCs - are present in membranes in trace amounts) and differs in various cell lines as well as in both membrane leaflets [6]. Moreover, the lipid composition of a biomembrane changes during carcinogenesis, and therefore normal and cancer cells differ in the content of the respective lipids (see e.g. [49]).

To compare the affinity of edelfosine to particular membrane lipids, the monolayers from this drug molecules mixed with compounds representing various groups of lipids were investigated.

Studying mixtures of edelfosine and cholesterol [50, 51], strong interactions have been found in monolayers. This may suggest that cholesterol can be of importance in edelfosine incorporation into cell membranes and its mode of action. The influence of the drug on cholesterol films were analyzed in a wide range of monolayer composition and at various surface pressures. It was concluded that the strongest interactions ( GExc ~ -2700 J/mol at 25 mN/m, 20o C) between both components occurred when edelfosine and cholesterol were mixed in 1:1 proportion [50, 51]. It has also been found that the addition of edelfosine fluidizes the film from cholesterol.

Regarding interactions with phosphoglycerides, the Langmuir monolayer experiments were focused on examining edelfosine

502 Anti-Cancer Agents in Medicinal Chemistry, 2014, Vol. 14, No. 4

effect on PCs and PEs, being the major lipids of the outer and inner layer of cellular membrane, respectively. Among PCs, three compounds differing in the degree of saturation of the chain(s) have been studied, namely 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) [52], 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) [50].

It has been found that the interactions between edelfosine and phospholipids are much weaker as compared to those with cholesterol. Such parameters of interaction, like A12 or GExc show either negative deviations from ideality (for edelfosine mixed with POPC or DOPC [50] or - in the case of mixtures with DPPCsmall

at room temperature, depending on the surface pressure region (1 mN/m and 25 mN/m, respectively) [52]. From the analysis of changes in the mean area (A12) values for the investigated edelfosine-PCs mixtures [50, 52] it seems that edelfosine molecules interact more strongly with unsaturated as compared to saturated phospholipids. It should also be mentioned that the unsaturation degree (i.e. one (POPC) versus two (DOPC) unsaturated chains) does not influence drastically the affinity of edelfosine to the phospholipid molecule [50]. To generalize, the strength of interactions between edelfosine and phosphatidylcholines is decreased in the following order: DOPC≈POPC>DPPC [50, 52].

The investigations on edelfosine/PC systems are performed much more frequently as compared to mixtures with phosphatidylethanolamines, which accumulate mainly in the inner leaflet of bilayer and thus serve as models of cytosolic shell of biomembrane. Three phosphatidylethanolamines differing in the structure of hydrophobic chains (DSPE, SOPE and DOPE), were investigated with edelfosine in mixed Langmuir monolayers [53]. It was evidenced that edelfosine molecules incorporate easily into the loosest PE layer, which is formed by DOPE molecules. On the contrary, the effect of edelfosine on DSPE (having fully saturated chains) as well as on mix-chained PE (SOPE) is highly unfavorable and provokes a phase separation between film molecules.

Edelfosine was also studied in mixed monolayers with sphingosine-containing lipids: egg sphingomyelin and ganglioside GM1 [54]. Experiments carried out for these mixtures indicated that edelfosine has a strong affinity to ganglioside ( GExc = - 2500 J/mol at 30 mN/m, 20o C), comparable to that of cholesterol, while its affinity to egg sphingomyelin is the weakest among all the investigated lipids ( GExc = +900 J/mol at 30 mN/m, 20o C) [54]. Further comprehensive studies performed for mixtures of edelfosine with synthetic sphingomyelin [55], involving Langmuir experiments complemented with Brewster angle microcopy and grazing incidence X-ray diffraction methods, evidenced that the incorporation of edelfosine into sphingomyelin film results in phase separation, decrease of films ordering as well as leads to the reorientation of sphingomyelin molecules at the air/water interface.

The affinity of edelfosine to the respective membrane lipids was attempted to be explained basing on the geometry of interacting molecules. This approach was applied when the reduction of hemolytic activity of edelfosine in liposomal formulation was analyzed. Busto [56] and Perkins [50] have found that the strongest decrease of hemolytic effect without a loss of activity is achieved when edelfosine is in combination with sterols and DOPE. The authors claimed that the ability of the latter lipids to decrease the hemolysis caused by edelfosine results from complementarity of molecular shapes of the above lipids and edelfosine. It was concluded that opposite molecular geometries: cone-shaped edelfosine (similarly to other lysophospholipids) and inverted coneshaped cholesterol and DOPE [57] ensure favorable packing of molecules in liposomal membrane, reducing in this way the drug release [50, 56]. In a set of similar experiments performed on liposomes composed from unsaturated phosphatidylcholines (POPC

Dynarowicz-Łątka and Hąc-Wydro

and DOPC), the observed effect for edelfosine was found to be much weaker. This was explained by the fact that PCs molecules are of truncated conical shape [57]; thus liposomes composed of POPC or DOPC do not meet a condition of geometrical complementarities with edelfosine.

It has been found that lowering of edelfosine toxicity by lipids can be correlated with the reduction of mean molecular area per lipid value and the strength of edelfosine-lipid interactions in monolayers. Namely, those lipids which were found to be most potent in the reduction of hemolytic activity of the drug (sterols, DOPE), in addition to geometrical complementarities, also cause the strongest reduction of the area per lipid values in monolayer experiments [50] and were found to interact strongly with edelfosine. For phosphatidylcholines this effect was much weaker (weaker reduction of area per lipid in mixed films with edelfosine and weaker interactions). This correlates well with less effective reduction of edelfosine toxicity in liposomes prepared from phosphatidylcholines, which do not meet optimal shapes complementarities with edelfosine [50]. Interestingly, these effects are stronger for unsaturated than for saturated PCs.

In general, the hypothesis that for a better reduction of toxicity the liposomal formulation should contain lipids of complementary structure to the drug, is in agreement with the results presented for sterols, PCs and PEs [50]. Moreover, the stronger reduction of the mean area per molecule for lipids mixed with edelfosine (as compared to the values resulting from the additivity rule), the better complementarity of molecular geometry and the reduction of hemolytic activity [50]. However, considering only the shape of interacting molecules, it is not clear why cholesterol is much more effective than DOPE, since both lipids are of similar shape. The authors [50] suggested that critical packing parameter, from which the molecular shape is estimated, does not consider favorable interactions between mixture components. The lack of correlation is also found when the strength of interactions of the drug with all the investigated lipids in binary films is compared. Based on the magnitude of deviations of the mean area per molecule from the values resulting from the additivity rule found for various edelfosine/lipid monolayers at the temperature of 20◦C [50-52, 54], the affinity of edelfosine to lipids decreases in the following order: cholesterol > GM1 > unsaturated PCs > saturated PCs > sphingomyelin. The most important question is why the interactions between edelfosine and ganglioside are only slightly weaker than those for edelfosine and cholesterol, since both ganglioside GM1 [58] and edelfosine are of similar conical shape? It seems that also in this case favorable interactions and separation of the negatively charged ganglioside by edelfosine prevail on the shape incomplementarity between both molecules. Thus, the geometrical complementarity should not be considered as a decisive factor responsible for the activity of edelfosine in membranes. In this context, the edelfosine-lipid interactions (which depend on the lipids structure and membrane composition) seem to be more important.

The results obtained from the experiments performed on simple binary lipid-drug monolayers allowed to draw some conclusions important for understanding of the mechanism of action of edelfosine and its selective effect on normal and tumor membranes (this will be discussed later). They have also outlined a further way of research, performed in more complex multicomponent monolayers.

2.2.2. Edelfosine in Multicomponent Lipids Monolayers

Studies on interactions between edelfosine and mixture of lipids in multicomponent monolayers were applied to explore issues, which may be directly correlated with the mechanism of action of edelfosine and its selectivity. These issues, namely the effect of edelfosine on model membranes of different fluidity, monolayers of various concentrations of cholesterol including raft-mimicking

Edelfosine in Membrane Environment - the Langmuir Monolayer Studies

system and role of gangliosides in edelfosine selectivity, were undertaken after a thorough analysis of the above-mentioned results performed for binary films.

2.2.2.1. Edelfosine in Mixed Monolayers of Various Fluidity - Studies on Ternary Films

Let us first analyze the investigations on the influence of edelfosine on mixed lipid monolayers differing in fluidity. This problem is directly related to the mechanism of selectivity of this drug since the experiments carried out on cell cultures have proved that the incorporation of edelfosine into tumor cell membranes is much easier as compared to normal cell membrane [43, 59]. Compositional differences of normal versus tumor cell membranes are responsible for distinct physical parameters of these membranes, especially their fluidity, which is higher in tumor membranes.

To study the influence of edelfosine on mixed monolayers differing in their fluidity, the following experiments were initially carried out [60]. Edelfosine, in various concentrations, was added into binary mixed films composed of phosphatidylcholine and cholesterol in two different proportions. One system modeled normal cell membrane and contained saturated PC (i.e. DPPC) and cholesterol. The other one, imitating tumor cell membrane, was of lower cholesterol amount and contained mix-chained PC (with one saturated and one unsaturated chain: POPC). Due to compositional differences, the foregoing mixed systems were of different condensation, namely normal model membranes were more condensed (less fluid) than tumor model system. The interactions found for normal membrane containing edelfosine (cholesterol- DPPC-edelfosine) were much stronger than those existing in the monolayer devoid of the drug molecules (cholesterol-DPPC). On the other hand, the addition of the drug to the tumor model membrane system was found to weaken the interactions existing between cholesterol and POPC molecules. Basing only on these results it cannot be resolved whether the drug molecules have favorable (from biological point of view) effect on one type of monolayer, while influencing negatively on the other one. However, it can be concluded that both tumor and normal model membranes are affected by edelfosine in a different manner. Moreover, it was found that the incorporation of low amount of edelfosine (up to 5% of monolayer components) practically does not affect the condensation of model normal membrane, while it strongly decreases the condensation of tumor model membrane. Upon further increase of edelfosine concentration, its effect on both model membranes becomes more pronounced, i.e. their condensation drastically decreases.

2.2.2.2. Cholesterol-dependent Influence of Edelfosine on Sphingomyelin/Cholesterol Monolayers - Edelfosine Disturbs Organization of Raft-like System

To analyze the effect of edelfosine on membrane domains (rafts) we have recently performed experiments on sphingomyelin/ cholesterol monolayers of raft-like sphingomyelin-to-cholesterol (SM/Chol) ratio = 2:1, containing various proportion of the drug [61]. Additionally, to investigate a correlation between the sterol content in the system and the effect of edelfosine, similar monolayer experiments were carried out on SM/Chol mixtures containing cholesterol in excess (SM:Chol=1:1 and 1:2). It was found that the incorporation of edelfosine into SM/Chol system, exactly of a raft-like proportion, destabilizes such a monolayer by weakening the SM-Chol interactions, thus exerting a fluidizing effect. However, when the SM/Chol system contains an excess of cholesterol, the effect of edelfosine is quite opposite, i.e. the interactions are strengthened and - in consequence - monolayers stability increases. Thus, it can be concluded that cholesterol present in excess “prevents” model membrane from destabilizing action of edelfosine. These results provide two major conclusions: a) strong affinity of edelfosine to cholesterol may contribute to the

Anti-Cancer Agents in Medicinal Chemistry, 2014, Vol. 14, No. 4 503

overall effect of the drug on model system, making the model membrane less susceptible to fluidizing and disordering effect of the drug; b) the organization of membrane rafts is strongly affected by edelfosine, proving that membrane domains may be important site of activity of this drug. It can be also concluded that the influence of edelfosine on model membranes varying in fluidity (the lower concentration of cholesterol, the more fluid is the monolayer) is different, which is consistent with the results presented in the previous paragraph.

This hypothesis is also consistent with previously discussed results for model tumor vs normal cell membranes. However, based only on findings published in the foregoing papers [60, 61] it is rather vague whether the influence of edelfosine on lipids membrane depends on films condensation per se or it is directly related to the level of cholesterol in the system. To clarify this issue, we have recently performed experiments on the influence of edelfosine on Chol/PC monolayers differing in their fluidity. The condensation of the studied model membranes was modified either by the content of cholesterol or by the structure (number of saturated chains) of phosphatidylcholine molecule at given cholesterol level (Chol/DSPC vs Chol/SOPC) [62]. These studies evidenced that there is no a simple correlation between monolayers fluidity and the effect of edelfosine on model system. Instead, it has been proved that the factor differentiating the influence of edelfosine on lipids monolayers is the level of cholesterol in the mixed film (the influence of edelfosine on model system is being strengthened with sterol concentration).

2.2.2.3. Gangliosides as Target Molecules for Edelfosine Incorporation in Membranes - Studies on Quaternary Monolayers

As it has already been mentioned, from the experiments for binary monolayers it is concluded that edelfosine molecules have strong affinity to gangliosides. This finding seems to be very important when investigating edelfosine mechanism of action and its selectivity since it is known that gangliosides are overexpressed in tumor progression [63, 64]. Therefore these molecules can be considered as tumor markers. It can be supposed that gangliosides may be directly related to the selective action of edelfosine on tumor membranes. This hypothesis was analyzed more deeply in Langmuir monolayer experiments [65]. The investigated films were composed of sphingomyelin and cholesterol mixed in the proportion found in glioblastoma cell lines. The same proportion corresponds to the average phospholipids/cholesterol ratio in brain tumors. The third component of the investigated films was ganglioside GM1, added in the proportion of 1, 5, 10 or 20%. Into these ternary monolayers (sphingomyelin/cholesterol/ganglioside) of different ganglioside concentration, edelfosine was incorporated in various and rather small amounts. It was found that at a very low concentration of edelfosine in the mixed system (1%), its influence on the model membrane was weak and completely independent on the concentration of the ganglioside. However, further increase of edelfosine content caused its incorporation into model membrane and augmented the strength of interactions in the mixed system as compared to those existing in the monolayer devoid of edelfosine. The effect of edelfosine was the more pronounced, the greater edelfosine concentration was in the system.

Considering the influence of ganglioside concentration, very interesting results were obtained. As it was mentioned earlier, at 1% of edelfosine in the mixed monolayer similar results were obtained for all the investigated systems, differing in the concentration of ganglioside. Thus, the content of ganglioside was found not to be important for edelfosine action on membrane when the drug was in such a low amount. However, when edelfosine content was increased to 5%, the results were similar only for these systems, which contained less than 10% of ganglioside, while at a higher ganglioside content (20%), the effect of edelfosine was much stronger. At 10% of edelfosine, differences between systems

504 Anti-Cancer Agents in Medicinal Chemistry, 2014, Vol. 14, No. 4

differing in the concentration of ganglioside become clearly visible and it can be observed that upon increasing of the ganglioside content, the effect of edelfosine becomes progressively stronger [65]. Thus, the correlation between the content of ganglioside and the effect of edelfosine is especially pronounced at a higher concentration of the drug.

3. LANGMUIR MONOLAYERS RESULTS VERSUS THE MODE OF ACTION AND SELECTIVITY OF EDELFOSINE

It is clear that in order to explain the mechanism of action of a drug, systematic and complex investigations on model and natural systems are required. These kind of multidirectional studies have also been carried out for edelfosine and other compounds belonging to this group of drugs. In this review we concentrate on the results derived from the Langmuir monolayer experiments, which were performed to verify the influence of edelfosine on model lipids membranes. In our opinion, the findings from these studies are important for planning of further research and in the future they may certainly help to clarify the mechanism of action of this drug. In this part of the review we summarize what kind of information on the edelfosine action provide the monolayer experiments. Below, the most important conclusions were compiled and the factors, which should be taken into consideration in further studies, were indicated.

3.1. Membrane Fluidity and Edelfosine Incorporation

The mechanism of action of edelfosine involves the penetration of drug molecules into the cell. The cellular membrane is thus the primary site of action for this drug. The normal and tumor membranes differ in their lipids composition, thus influencing the physical parameters of membranes. The elevated proportion of unsaturated chains and decreased concentration of cholesterol (in most of edelfosine-sensitive tumor cell lines) make the membrane of tumor cell more fluid as compared to normal membranes [66, 67]. It has been found that edelfosine molecules do not incorporate into normal cell membranes, while they easily penetrate membranes of various tumor cells. Hence, the conclusion is that the membrane of normal cell acts as a natural barrier for the incorporation of edelfosine molecules [43, 59].

The foregoing easier incorporation of edelfosine into more fluid membrane reflects also in the Langmuir monolayer studies. Edelfosine was found to have a stronger affinity to model membranes (Langmuir films) formed by unsaturated lipids (e.g. POPC, DOPC, DOPE) as compared to their saturated analog (DPPC). As it was found in several experiments, unsaturated lipids form monolayers of lower condensation (more fluid) than those possessing fully saturated chains [68]. Thus, comparing the monolayers formed by lipids differing only in their saturation degree, it is evident that edelfosine interacts more strongly with less condensed films. Interestingly, stronger affinity to less condensed model membranes (Langmuir films) was found also for another antitumor lipid - miltefosine (hexadecylphosphocholine), belonging to the group of alkylphosphocholines (APCs) [69]. The experiments were based on injecting of miltefosine solution underneath a phospholipid monolayer, compressed to high surface pressure (corresponding to the condition in natural membrane), and afterwards the variation of the mean molecular area of lipid molecules at a constant surface pressure value was monitored. In another set of experiments, the variations of the surface pressure were recorded at a constant surface area [69]. Such experiments were carried out for several phospholipids monolayers of various condensations, namely: DPPE, DPPG, DPPC (condensed films) and POPE and POPC (fluid films). It was concluded that miltefosine cannot insert into condensed monolayers, while it becomes easily incorporated into fluid films.

The effect of edelfosine on monolayers of a different physical state (condensation) was also investigated in experiments

Dynarowicz-Łątka and Hąc-Wydro

performed on binary mixed cholesterol/PC films [60]. It should be, however, pointed out here that in these experiments the condensation of the film was modulated by both the degree of the saturation of phospholipids (POPC versus DPPC) as well as by the concentration of cholesterol in the mixture. It was found that the effect of edelfosine on model systems differing in their condensation degree is different, namely the addition of edelfosine into less condensed film was thermodynamically unfavorable on model membrane, as it provoked a weakening of interactions in the mixture as compared to those existing in the monolayer devoid of edelfosine, and even at a lower concentration edelfosine changed the molecular organization of more fluid (tumor) membrane, i.e. decreased the condensation of the film. However, when analyzing these results it is clear that not only the monolayer state but also the sterol content in the mixed system should be taken into consideration. Since it was proved that edelfosine shows strong affinity to cholesterol, it is possible that the effect observed for the investigated cholesterol/PC monolayers is determined not only by the state of the film but also by the sterol concentration. In the foregoing investigations the cholesterol/DPPC films were of higher proportion of cholesterol than cholesterol/POPC monolayers. It is thus possible that the observed stabilizing effect of edelfosine incorporated into mixed monolayers of higher sterol content results from strong and thermodynamically favorable edelfosinecholesterol interactions.

To get a deeper insight into this problem, systematic experiments have been performed aimed at studying the relationship between the condensation of the monolayer and the effect of the ether lipid on the model system [62]. These investigations were carried out on Chol/PC film varying in the fluidity regulated either by the content of cholesterol or by the structure of PC molecule (DSPC vs SOPC) at a given cholesterol content. It was found that the influence of edelfosine on the studied system is strongly determined by the level of cholesterol in the mixed Chol/PC monolayer. Although the saturation of PC molecules in mixtures at a given sterol concentration also influenced - to some degree - the interactions of edelfosine with model membranes, the impact of cholesterol concentration was found to be the most decisive factor.

The foregoing findings obtained from Langmuir monolayer experiments were of much help in understanding differences in sensitivity of various cells to edelfosine. An excellent example is the fact that in biological studies edelfosine was found not to affect normal cells, while it easily incorporates into more fluid tumor cells. Since normal vs tumor membranes differ in the level of cholesterol (which is a regulator of membrane fluidity) it can be proposed that the membrane composition is undoubtedly responsible for some kind of “natural selectivity” of edelfosine.

3.2. Cholesterol Role and Membrane Rafts

Edelfosine and other compounds belonging to ALPs are characterized by high selectivity accompanied by low toxicity. This is related to the above mentioned tendency of favorable incorporation of the drug into tumor membranes, which - in turn - is facilitated by their lower fluidity regulated e.g. by the level of cholesterol as compared to normal membranes. However, it is possible that these differences in membrane properties are only one of the factors responsible for this drug selectivity. Since the composition of cellular membrane changes upon cancer progression, some specific components of a membrane may also be responsible for a stronger affinity of edelfosine to tumor cell membranes. Based on the published results obtained from the Langmuir films experiments and other membrane mimicking-systems it is possible to analyze which membrane lipid may be considered as a target for edelfosine and which molecule can play a specific role in the mechanism of selectivity of this drug.