Evaluation of the In Vitro Stability of Stimuli-Sensitive Fatty Acid-Based Microparticles for the Treatment of Lung Cancer

The fatty acid-based microparticles containing iron oxide nanoparticles and paclitaxel (PAX) are a viable proposition for the treatment of lung cancer. The microparticles inhaled as a dry powder can be guided to selected locations using an external magnetic field, and when accumulated there, the active compound release can be triggered by local hyperthermia. However, this general strategy requires that the active compound is released from microparticles and can reach the targeted cells before microparticles are removed. Isothermal titration calorimetry was used to demonstrate that the components of microparticles were released and transferred to albumins and lipid bilayers. The morphology of the measured particulates was studied with scanning electron microscopy and dynamic light scattering. To determine the cytotoxicity of microparticles, cell culture studies were done. It has been shown that the transfer efficiency depends predominantly on the fatty acid composition of microparticles, which, together with the active ingredient, accumulate predominantly in membrane structures after being released from microparticles and before entering the cytoplasm. The release process is sufficient; hence, paclitaxel-loaded microparticles effectively suppressed the proliferation of A549 human lung epithelial cells of malignant origin (IC50 values for both lauric acid-based and myristic/palmitic-based microparticles containing paclitaxel were below 0.375 μg/mL), while reference microparticles were noncytotoxic.


■ INTRODUCTION
Lung cancer is a leading cause of cancer-related death. 1,2 Due to the fact that in many cases this disease is diagnosed at the late stage and is more prevalent in elderly people, therefore very few options of radical treatments are available. 3 Thus, there is an ongoing search for new effective treatments of lung cancer. One of the possible approaches is the application of a targeted drug delivery system with the potential to enhance the efficacy simultaneously reducing side effects. 4,5 There are numerous works testing the viability of such approach; however, most of them do not have the capacity for the localized accumulation of the drug carrier in affected regions of lungs. One possibility to overcome this limitation is the construction of the delivery system containing, in addition to the active ingredient, the superparamagnetic iron oxide nanoparticles (SPIONs). 6 Nanoparticles can be guided by an external magnetic field with the support of an imaging infrastructure. In addition, NPs can be used to induce local hyperthermia, and therefore serve as a trigger for the release of the active ingredient. 7 This strategy becomes effective only when the carrier satisfies certain requirements. First, the lung anatomy and physiology impose the size limits for the particulate formulation. The drug carrier needs to be in the range between 1 and 10 μm so it can reach and remain in the alveoli. 4 When deposited in the alveoli, the microparticle (MP) is immersed in the complex matrix of mucus from where it might be internalized by macrophages or remains there, releasing the active ingredient locally. 8,9 This scenario is even more probable in stressed lungs when both mucus and underlying cells immersed in the extracellular matrix are severely altered. 10−12 It is believed that the efficiency of drug release can be potentially enhanced by changing the particulate state usually by local hyperthermia. An example of such drug delivery vehicle is a solid microparticle (MP) formed from fatty acids, loaded with NPs and anticancer drug paclitaxel (PAX). 13,14 The MPs are delivered to a patient via inhalation using a standard dry powder inhaler. Upon inhalation, the MPs can be guided to the tumor site using an external magnetic field. Once the MPs are accumulated at the tumor location, the alternating electromagnetic field is applied. NPs embedded in the microparticles heat up to around 42−47°C, resulting in the melting of the fatty acid matrix of the MPs. This may accelerate the active ingredient PAX release at the tumor vicinity. After treatment, the particulate remnants are absorbed by surrounding tissues and/or removed from the lung via natural clearance mechanisms. Consequently, the delivery system should conform to a specific kinetic requirement dependent on lung physiology and the transfer of active ingredients between microparticles and malignant cells.
Experimental results regarding effects of microparticles components on their functions and properties were published elsewhere. SPIONs were tested with respect to their cytotoxicity and modified with silica layers to enhance their biocompatibility. 15 In addition, it has been shown that there is a weak effect of modified SPIONs at relevant concentrations on physicochemical and topological properties of fatty acid microparticles. 16 As presented elsewhere, experiments performed on cell cultures demonstrated that microparticles containing PAX are much more effective against a cancer cell as compared to those without it. In addition, it has also been shown that the effect of temperature on efficacy of microparticles is much higher when they are formed from lauric acid (LAU). No such effect has been observed for microparticles formed from a mixture of myristic and palmitic acids. In other studies, it has been observed that saturated fatty acids alone can reduce malignant cells (A540 line) viability with no effect on nonmalignant cells (BEAS-2B line). 17 Therefore, there are experimental evidences that both PAX and saturated fatty acids affect malignant cells, whereas SPIONs do not affect microparticles efficacy, properties, or topology. As stated above, when microparticles are inhaled, they first encounter the mucus layer. 18 Consequently, the next issue that needs to be addressed is the stability of microparticles in mucus, where microparticles integrity may change due to the release of their components in the surrounding milieu.
The molecular mechanism of a hydrophobic compound release from a microparticle is not a straightforward process. All pharmacologically active ingredients, regardless of their chemical nature, are weakly hydrophobic (typically log P > 2, where log P is the logarithm of the octanol/water partition coefficient). 19 Due to the hydrophobicity of microparticle's components is the reason that microparticles cannot be simply solubilized in aqueous media. Consequently, when in mucus, the hydrophobic compound likely remains in the particulate matrix or associates with surrounding proteins (albumins) and/or biological membranes. The changing phase of the fatty acids, when induced by the elevated temperature, may alter the active compound's affinity toward the microparticle, thereby enhancing the release. However, this would not change the log P of the compound, so its affinity toward proteins and biological membranes remains a critical factor in the release process. 20 The other possible mode of action is that the microparticles are internalized entirely by macrophages and/or cancer cells providing the additional boost to the treatment. 21 In this paper, the thermodynamics of the active ingredient (PAX) and/or saturated fatty acid release from fatty acid microparticles containing NPs in the environment mimicking the conditions in situ is presented. Moreover, in vitro efficacy of the system in contact with malignant lung epithelial cells is shown. Fabrication of Fatty Acid-Based Microparticles (MPs). The MPs were prepared based on LAU alone or a eutectic mixture of MYR/PAL. Two types of MPs based on each fatty acid were preparedempty MPs and NP + PAX-loaded MPs.

■ EXPERIMENTAL SECTION
Surface Modification of NP. Iron(III,IV) oxide nanoparticles (NP) were surface-modified with a mesoporous silica layer to inhibit the dissolution of the nanoparticles and iron release according to the methods described earlier. 22,23 Briefly, 1 g of the NPs was mixed with 175 mL of ethanol/UHQ/ammonia mixture (volumetric ratio: 32:8:1) and homogenized using an ultrasound probe for 15 min. Simultaneously, 1 g of CTAB was dissolved in 25 mL of the abovementioned ethanol/UHQ water/ammonia mixture and added to a NP suspension. After 15 min of sonication, 1 mL of TEOS was added and sonicated for additional 5 min. The obtained suspension was transferred into falcon tubes and vigorously shaken using a horizontal shaker overnight. Modified NPs were purified via repeated centrifugation and rinsed with ethanol (3 mL × 25 mL) and water (3 mL × 25 mL) followed by air drying at 60°C for at least 12 h.
Preparation of MYR/PAL Mixture. MYR (5.8 g) and PAL (4.2 g) were mixed and heated to 80°C for 30 min. The obtained mixture was homogenized using a vortex mixer for 3 min and cooled to room temperature.
Fabrication of MPs. LAU and MYR/PAL-based MPs were fabricated using the hot oil-in-water emulsification method. Empty MPs were prepared by melting 200 mg of fatty acid in a water bath at 65°C for LAU and 69°C for MYR/PAL (20°C above T m of fatty acids). Melted fatty acid was poured to a falcon tube containing 2 mL 10% w/w PVA heated to the same temperature as the fatty acid. The tube was vortexed vigorously for 90 s, and the obtained emulsion was immediately poured to a vial containing around 30 mL of liquid nitrogen. The MPs were purified via repeated centrifugation (12 000 rpm, 10 min, 4°C) and rinsed with UHQ water. The MPs were frozen at −80°C and freeze-dried for 24 h (Labconco, FreeZone 6 ). NP-loaded and NP + PAX-loaded MPs were prepared in a similar way. Ten milligrams of NP (5% w/w) and 10 mg of PAX (5% w/w) were added to melted fatty acids and homogenized using an ultrasound probe (2 min, 40% amplitude, pulse mode: 10 s ON, 5 s OFF, Vibra-Cell, Sonic & Materials, Inc). They were further processed according to the protocol described above.
Physicochemical Characterization of the MPs. Microparticles Surface Change Analysis. For the measurement of the surface ζ potential of MPs, the MPs were suspended in UHQ water at about 1 mg/mL concentration and poured into dedicated Omega cuvettes (Anton Paar, Austria). The measurements were performed in triplicates for each type of MPs (at least 50 runs for each measurement) using LiteSizer 500 (Anton Paar, Austria).
Scanning Electron Microscopy (SEM) and Average Size Determination. NP + PAX-loaded MPs were observed using a scanning electron microscope (Zeiss Ultra plus, Zeiss). The MPs were dispersed onto carbon tape and coated with gold. SEM imaging was performed at 2 keV. The average size of the MPs was determined using the standard image analysis tools.
Determination of Phase Transition Temperature. Approximately, 3.5 mg of the MPs were weighed in aluminum crucibles and closed with pierced lids. Differential scanning calorimetry (DSC) measurements were performed using DSC 1 (Mettler Toledo) within a temperature range of 0−100°C, at a 10°C/min heating rate and with a 30 mL/min N 2 flow rate. Data analysis and determination of the melting temperatures (T m ) as well as normalized heat of fusion of the MPs were conducted using STARe software.
In Vitro Release of PAX from MPs Determined with Dialysis. The studies on PAX release from the MPs were performed using Micro-Float-A-Lyzer chambers (MWCO = 50 kDa, Spectrum Labs). Langmuir pubs.acs.org/Langmuir Article LAU + NP + PAX or MYR/PAL + NP + PAX (100 mg) was suspended in 1 mL of the following acceptor fluids: 5% aqueous suspension of soybean lecithin (Phospholipon 90 Lipoid AG) liposomes dissolved in PBS buffer and 4% aqueous solution of bovine serum albumin (Sigma-Aldrich) in PBS buffer. The first acceptor fluid was used as a model of a cell membrane, while the second one was supposed to imitate the extracellular environment and mucus. Particles following deposition in the alveoli are entangled in mucus; therefore, the main route of hydrophobic active substance transfer is the release from particulates and subsequent reabsorption on extracellular and cellular structures. The process depends predominantly on the hydrophobicity of compounds forming the delivery vehicle. 12,18 The experimental models used are intended to evaluate a spontaneous transfer of amphiphiles from the particulates to albumin, which represents protein components of mucus, 24 and the lipid bilayer, which models biological membranes. 25 Specifically, the chambers were filled with 650 μL of the MP suspension and immersed in 70 mL of the acceptor fluid. The studies were performed at 37 and 45°C for 24 h, and the chambers were continuously stirred at 100 rpm. Samples (500 μL) of the external acceptor fluid were taken at predetermined time points (1, 2, 4, 8, 24 h). PAX was extracted from the acceptor fluid using methanol (high-performance liquid chromatography (HPLC) grade, Sigma-Aldrich). Specifically, 200 μL of the sample was mixed with 1.8 mL of methanol, vortexed for 30 s and centrifuged (5000 rpm, 5 min, 21°C) to remove precipitates. The concentration of PAX was measured in triplicates using HPLC, as described below.
Determination of PAX Concentration. The concentration of PAX encapsulated in the MPs was determined using high-performance liquid chromatography (HPLC) coupled with a variable wavelength UV−vis detector (Knauer) based on the modified Agilent 5988−7973 application note. Five milligrams of NP + PAX-loaded MPs was dissolved in 5 mL of methanol. The solution was sonicated in an ultrasonic bath at 50°C for 30 min to ensure complete dissolution of PAX from the MPs. NPs were removed from the solutions via centrifugation (6000 rpm, 5 min, 25°C) and the resulting supernatants were transferred to the amber glass sample vials (VWR) for analysis. Twenty microliters of the sample was injected on an alkyl C18 column (4 mm × 150 mm) at 25°C. The mobile phase was composed of the following: Awater and Bacetonitrile. The flow rate was adjusted to 0.8 mL/min with the following mobile phase gradient: at 0 min 50% B, at 10 min 90% B, wash at 12 min 50% B, stop at 17 min. All samples were assayed at 204 nm. The concentration of PAX was measured in triplicates. Samples containing lipid vesicles (200 mg) were mixed with 1800 mg of methanol (HPLC grade) prior to the measurement. Samples containing albumin (200 mg) were mixed with 1800 mg of acetonitrile followed by centrifugation (5000 rpm) for 5 min. The resulting supernatant was filtered through a 0.22 um filter and evaluated for the presence of PAX.
Preparation and Characterization of Liposomes. Multilamellar DOPC vesicles (MLVs) were prepared by the dry film method. In short, lipid was dissolved in chloroform. The organic solvent was removed by the stream of argon, and the residues of chloroform were removed during the storage under low pressure. The obtain dry lipid film was hydrated overnight in PBS (137 mM NaCl, 2.7 mM KCl, and 10 mM phosphate buffer) at 25°C. The final concentration of lipid in the sample was adjusted to 40 mg/mL. Next, the suspension of MLV liposomes was extruded through a 100 nm polycarbonate membrane (Nuclepore Corp.). The extrusion was carried out using an automated mechanical extruder (Lipid Systems, Poland). The size distribution and ζ potential of vesicles were determined using the dynamic light scattering (DLS) method (Zetasizer Nano ZS, Malvern, U.K.). Prior to the measurement, the samples were diluted 50-fold and the buffer was filtered through a membrane with 0.22 μm pores (VWR). Using the same device, ζ potential measurements were carried out. Samples were diluted in 10 mM KCl and put into a folded capillary ζ potential cell (Malvren, U.K.).

Microparticle Interaction with the Lipid Membrane and
Albumins. In the experiment, the aqueous suspension of microparticles in PBS buffer (concentration is equal to 0.4 mg/mL) was placed in the mixing chamber of the isothermal titration calorimeter (ITC) (NanoITC, TA Instruments). Then, the liposomes (lipid concentration 0.5 mg/mL) or albumin (10 mg/mL) suspension in PBS buffer was titrated in a chamber. In the reversed setup, the titration of liposomes or albumin with microparticles was not practical due to their rapid sedimentation in the syringe. All measurements were carried out at 37 or 45°C, the volume of the injections was set to 10 μL, and the speed of mixing rotor to 350 rpm. To eliminate the possible influence of titrant diffusion into the chamber, the first injection of a small volume (1.14 μL) was discarded.
Determination of In Vitro Efficacy of the MPs. The MPs were tested in contact with malignant lung epithelial cells (A549, ATCC CCL-185TM). Cell culture was performed using Dulbecco's modified Eagle's medium (DMEM, PAN Biotech, Germany) supplemented with 1% fetal bovine serum (FBS, South America Origin, Pan Biotech, Germany) and 1% penicillin/streptomycin (PAN Biotech, Germany) at 37°C, 5% CO 2 , and a humidified atmosphere. The cells were seeded at 5 × 10 3 cells/well in a 96-well plate and cultured for 24 h. Unloaded and NP + PAX-loaded MPs based on LAU and MYR/PAL were UV-irradiated for 30 min and suspended in a cell culture medium at concentrations ranging from 0.125 to 5 μg/mL. After 24 h of seeding, the cell culture medium was withdrawn from the well plate and replaced with an MP-containing medium. Control cells were incubated in DMEM without the addition of MPs (MPs concentration: 0 μg/mL). The cells were cultured in the presence of the MPs for 24 h. The experiment was performed in triplicate for each type of the MPs.
Cell viability was evaluated using the metabolic activity test (resazurin reduction assay). After 24 h of addition of the MPs, the cell culture medium was removed and 100 μL of fresh DMEM containing 5% (v/v) of AlamarBlure reagent (resazurin sodium salt, 0.1 mg/mL dissolved in PBS, Sigma-Aldrich) was added to the wells. After 3 h of incubation, 80 μL of the cell culture medium was transferred into a black 96-well plate and the fluorescence was measured at λ ex = 530 nm, λ em = 590 nm using a microplate reader (FluoroSTAR Omega, BMG Labtech, Germany). Resazurin reduction percentage was calculated as follows where F x is the fluorescence of the sample, F 0% is the fluorescence of DMEM with AlamarBlue reagent without cells, and F 100% is the fluorescence of the completely reduced reagent (DMEM with the reagent was autoclaved for 15 min at 121°C). IC 50 values for all types of MPs were determined based on the percentage reduction of resazurin. Live/dead fluorescence staining was also performed 24 h after the addition of MPs. In brief, the cell culture medium was removed from the wells and replaced with 100 μL of FluoroBrite DMEM (Gibco, Life Technologies) containing 0.1% calcein-AM (Sigma-Aldrich) and 0.1% propidium iodide (Sigma-Aldrich). After 20 min of incubation, the cells were gently washed with 100 μL of PBS and visualized at 100× magnification using a fluorescence microscope (Axiovert 40 CFL with HXP 120 C Metal Halide Illuminator, Zeiss, Germany).

■ RESULTS AND DISCUSSION
The prepared microparticles (MPs) were spherical and had a diameter in the range of 1.9−3.6 μm. Their ζ potential varied from −9.9 ± 0. 7 to −12.5 ± 0.7 mV and −15.3 ± 0.9 to −16.5 ± 0.7 mV for LAU and MYR/PAL-based MPs, respectively. A variation in MP composition affects their surface topology, as shown in Figure 1A. The phase transition temperatures for LAU and MYR/PAL particles were 45 and 49°C, respectively, and were little affected by the addition of NP and PAX ( Figure  1B and Table 1). In both LAU and MYR/PAL-based MPs, Langmuir pubs.acs.org/Langmuir Article normalized heat of fusion was significantly lower for NP + PAX-loaded MPs. This phenomenon is attributed to the decreased content of fatty acids, as the additives such as NP and PAX constituted up to 10% of the MPs mass. All MPs were prepared from highly hydrophobic compounds, namely, LAU (log P ∼ 4.6), MYR (log P ∼ 6.11), PAL (log P ∼ 7.17), and PAX (log P ∼ 3.66). log P can be used as an indicator of a compound propensity for the exchange between supramolecular aggregates. The compound with a low log P value can exchange between microparticle and biological structures (membranes or globular proteins). 18,26,27 To measure the release of the hydrophobic/amphiphilic compound from MPs, the aqueous incubation medium is not sufficient, so aqueous suspensions of albumins or liposomes were used instead. 20,28−31 The dialysis of MPs in the aqueous medium containing albumins or liposomes showed that the PAX released from microparticles was below the detection limit. No PAX was detected in the receiving solution, regardless of the experiment duration or the composition of the receiving medium. This result may indicate that the MPs are stable in physiological fluids. The major intrinsic uncertainty of such experimental system is erased from the equilibration of the hydrophobic (amphiphilic) compound within the hydrophilic membrane. 31 The best way to overcome this limitation is the application of a method that does not require the separation of acceptors (liposomes or albumins) from MPs and it is preferable label free. The isothermal titration calorimetry fulfills these requirements; the reactants are mixed in a single volume and the heat flow produced by the accompanying processes is measured. 32,33 The titration experiments were performed at two temperatures: 37°C (normal physiological temperature) and 4°C (intended temperature as induced locally by hyperthermia). Figure 2 shows examples of thermograms obtained at 37°C from MPs formed from LAU and MYR/PAL. Interestingly, the enthalpy changes during the dilution of MPs formed from LAU are positive, whereas those for MPs formed from MYR/PAL are negative. Figure 3 shows examples of cumulative enthalpies    Langmuir pubs.acs.org/Langmuir Article constructed from enthalpies released by MPs diluted with buffer containing albumin ( Figure 3A) or liposomes ( Figure  3B) and corrected for the dilution of a solution containing MPs with a titrant as well as the dilution of the titrant itself (data not shown). It shows that the plots of a cumulative enthalpy as a function of albumin or liposome concentrations (mg/mL) are both quantitatively and qualitatively different. First, the energy released during the dilution of MPs (calculated per unit of weight of acceptor) with liposomes is much higher than the energy released when albumin is used (by more than an order of magnitude). Furthermore, thermograms in the two cases are qualitatively different. When albumin is used as a titrant, the energy released during a single injection remains unchanged within concentration limits used in the experiment (see Figure 2, for example, of thermograms). When liposomes are used, the energy released in subsequent injections changes even though the amount of the lipid in the titrant is much less. These differences may reflect the weak interaction of a component(s) forming MPs with albumin. The reduced amount of energy that accompanies fatty acid binding with albumin can be explained by weak interaction and/or a limited number of binding seats on the protein. 26 No such limitation exists when the lipid bilayer is used as an acceptor. 31,34 The rate of the enthalpy change, as a function of the quantity of an acceptor (mg/mL), is determined and compared for different MPs at two temperatures ( Figure 4).
The energy released during the dilution with albumin or liposomes is not significantly different for two selected temperatures and all types of MPs. Therefore, the presented results show that the application of local hyperthermia may not be very effective for highly hydrophobic compound release. 35−37 The absolute energy released when MPs formed from LAU were titrated with albumin is higher than that released when MPs were formed from MYR/PAL fatty acids. This may result from the difference in the energy barriers for fatty acids and/or PAX when leaving microparticles and/or difference in a number of binding seats on albumins. The limited number of binding seats on albumins or a lower compound association strength with MPs explains the effect despite the fact that the association constant of MYR/PAL with albumin is much higher than that for LAU. 31 When MPs were titrated with lipids, the number of binding seats on liposomes is similar for both fatty acids; therefore, the higher affinity of MYR/PAL to the membrane is now more prominent. 30 When PAX is present in the microparticles, the transfer energy is always lower than for MPs formed from fatty acids alone.
The quantity of energy released (absorbed) by the system to reach a new equilibrium is determined by the isothermal titration calorimetry. 32,34 However, the isothermal titration calorimetry can also be used to follow the kinetics of system equilibration. A thermogram can be conveniently used for the characterization of a well-defined and relatively simple processes. 38 In a complex system, the titrant may trigger a sequence of events, as is the case in the presented studies. Specifically, upon titration, a compound may dissociate from MPs followed by its association with a protein or a lipid bilayer. The association with the lipid bilayer by itself is a complex sequence of events such as the association with the outer surface of the liposome followed by the transfer to the inner surface. 34 To make the situation even more complicated, there are at least two compounds in the presented MPs capable of crossing the aqueous barrier between two hydrophobic particulates (MPs and albumins or liposomes). Some properties of the system are reflected in the dependence of enthalpy released on a number of injections. 39,40 Additional information can be extracted from the evolution of the rate of the heat flow following a single injection as a function of time. 41 Figures S1 and S2 (see the Supporting information) show examples of energy flow following a single injection of 10 mg/mL albumin solution or 0.5 mg/mL unilamellar lipid vesicles formed from DOPC to microparticles formed from fatty acids alone or from fatty acids mixed with NP and PAX. In addition to the differences mentioned before regarding the direction of energy flux, there are additional, less obvious differences. For example, when PAX is present in microparticles, thermograms are qualitatively different, as illustrated by plots showing the time correlation of microparticles with and without PAX ( Figure 5). The difference is more prominent when particles are formed from LAU. When microparticles are titrated with liposomes, the presence of PAX dramatically alters the shape of the rate of the heat flow following a single injection. The effect of PAX on the rate of the heat flow when MPs are formed from a MYR/ PAL mixture is qualitatively different. The difference between MPs formed from different fatty acids can be interpreted in terms of the partition coefficient. The LAU has a relatively low value of the partition coefficient; hence, its stability can be affected more by the presence of PAX. No such effect is observed when MPs are formed from highly hydrophobic, and  Consequently, the correlation between microparticles with and without PAX is similar. Thermograms can be quantitated using different methods and approaches. 42 One of the simplest approaches is to compare the duration of the equilibration process following a single injection, as summarized in Figure 6.
Here, the picks are compared based on their width at the halfhigh. From data presented in Figure 6, it is clear that MPs formed from MYR/PAL equilibrate much slower than samples containing microparticles formed from LAU. The effect is especially prominent when MPs are titrated with liposomes. This observation is consistent with previous results, indicating that the transfer of highly hydrophobic MYR/PAL is much slower than that of the less hydrophobic LAU. The exposure of liposomes to microparticles results in the transfer of matter, which should be reflected by the change in the liposome and microparticle properties. Table 2 shows that after the incubation of liposomes with microparticles for 60 min at 40°C, both their size and ζ potential change. The average liposome size increases from 118 ± 1 nm to about 130 nm when they were exposed to fatty acids alone and up to about 142 nm when exposed to MPs with PAX. At the same time, the ζ potential decreased from −2 mV to about −30 mV, regardless of the type of microparticles used. The decreased surface potential shows that negatively charged fatty acids are transferred to liposomes and generates the electrostatic surface potential. The increase in the liposome size is more prominent when PAX is present in the microparticles showing that, in addition to fatty acids, neutral PAX may also be transferred to liposomes. 29 Regardless of the in vitro release trials, the efficacy of the MPs was tested in contact with malignant human lung epithelial cells (A549 cell line). Unloaded MPs did not affect the viability of A549 cells, even at the highest tested concentration (5 μg/mL) ( Figure 7A). In the case of NP + PAX-loaded MPs, the first significant deterioration in cell viability was observed for concentrations as low as 0.125 μg/ mL. Further increase in the MP content resulted in more    Figure 7B). A549 cells cultured with 5 μg/mL of unloaded MPs were viable, with typical morphology and only less than 2% of dead cells were observed. In the case of A549 cells incubated with 5 μg/mL of NP + PAX-loaded MPs, scarce live cells were found. Remaining live cells were poorly spread and spherical, indicating early apoptosis. Interestingly, a low number of dead cells were also observed in those samples; however, it was related to the fact that the majority of them were removed during the staining procedure.

■ SUMMARY AND CONCLUSIONS
Fatty acid microparticles equipped with magnetically active nanoparticles are considered as an effective tool for the treatment of lung cancer, enabling a targeted delivery and triggered the release of an active compound from microparticles at the selected location. The role of the magnetic nanoparticles is to guide microparticles toward their destination and trigger the release of an active ingredient using hyperthermia induced by an external magnetic field. To be effective, such device should be able to release its cargo, when in the physiological fluid on the surface of an alveoli. The release exclusively depends on physiochemistry and biophysics of microparticles and the properties of an active ingredient and surrounding biological matrix. In this paper, the dependence of the release process of paclitaxel from fatty acid microparticles on the type of fatty acid used, temperature, and the presence of two elementary biological structures (albumin and lipid bilayers) was investigated. It was determined, using isothermal titration calorimetry, that the fatty acid used to form microparticles greatly influence the thermodynamics of the releasing process. When microparticles were exposed to albumins or lipid bilayers, the desorption of fatty acids and paclitaxel was detected and the release profile depends strongly on the fatty acids used, showing that the composition of microparticles is an important parameter in designing the targeted drug delivery system. Surprisingly, the effect of temperature was small, showing that the application of hyperthermia for releasing the hydrophobic compounds from microparticles may not be as effective as assumed. Nevertheless, the MPs loaded with 5% PAX effectively suppressed the growth of malignant lung epithelial cells (A549) even at concentrations as low as 0.125 μg/mL, while unloaded MPs were not cytotoxic for cells. The performed studies showed that fatty acid-based MPs loaded with magnetic nanoparticles and the anticancer drug, paclitaxel, are promising materials for the localized treatment of lung cancer.
■ ASSOCIATED CONTENT * sı Supporting Information The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.langmuir.0c02141. Thermogram of a single 10 μL injection of 10 mg/mL albumin (two left panels) and liposomes (two right panels) to LAU and LAU+NP+PAX microparticles ( Figure S1); the thermogram of a single 10 μL injection of 10 mg/mL albumin (two left panels) and liposomes (two right panels) to MYR/PAL and MYR/PAL + NP + PAX microparticles ( Figure S2); thermograms related to the correlation data in Figure 5 (PDF)