An important requirement for new drug delivery carriers is their nontoxicity. Therefore, initial studies concerned cytotoxicity of the nanoparticles on cells. Representative results for HMEC (human microvascular endothelial cells) are shown in
Figure 1 and in
Table 1,
Table 2,
Table 3 and
Table 4. Although a significant decrease in cell viability is observed, a cell viability above 80% is maintained under all but the highest concentrations tested (25 µM and 50% of particle suspension). This suggests only a mild cytotoxic effect of the particles on the tested cells. On the other hand, they can enter a cell confirmed by confocal microscope pictures. As the nanoparticles satisfied the necessary condition, further analysis was carried out by the ESR method.
2.1. Effect of Temperature
The ESR spectrum of the nanoparticles with attached spin labels had the form shown in
Figure 2 and
Figure 3a.
Figure 2 illustrates a typical ESR spectrum in the whole range of magnetic field. The broad line (ΔH = 84mT, g = 2.220) is derived from magnetite core with superparamagnetic properties, while a signal in the middle of the spectrum (enlarged in
Figure 3a) is typical of a spin label.
The ESR spectra of the nanoparticles studied in suspensions without (
Figure 3a) and with (
Figure 3c) cells are characterized by similar structure and spectroscopic parameters, such as peak-to-peak line width, hyperfine splitting, and spectroscopic splitting factor. At room temperature (293 K), the molecular dynamics of magnetite core and spin labels are very high in the selected concentration range. Therefore, interaction anisotropy of the nanoparticles with the environment is largely averaged, and the differences in the values of spectroscopic parameters are within the limits of measurement uncertainty. In our previous studies [
38], we clearly demonstrated that ESR spectra differentiated in lower temperatures of ESR measurements depending on environmental conditions. The differences were observed for the samples in various environments, including nanoparticles inside and outside cells. Finally, the measurement temperature of 240 K was chosen as an optimal value for ESR spectra recording of samples with cells incubated previously as standard at 310 K, because it guaranteed the greatest changes in the structure and spectroscopic parameters. In the selected conditions, the interactions of nanoparticles in a different environment, especially in the presence of cells, varied significantly (
Figure 3b,d).
Figure 3 clearly demonstrates that ESR spectra of the nanoparticles studied without (
Figure 3a,e,g) and with cancer cells (
Figure 3c) are very similar in room temperature. At the same time, those recorded at 240 K are significantly different (
Figure 3b,d). The differences observed are an effect of interaction and/or attachment of the nanoparticles to the cancer cells.
The differentiation of ESR spectrum structure recorded at 240 K was observed in the presence of cancer cells (
Figure 3d), as well as human microvascular endothelial cells (
Figure 3f) and yeast cells (
Figure 3h) incubated previously at 310 K. Therefore, yeast cells were used as models for the analysis of interactions between nanoparticles and cells, including endocytosis and determination of changes in ESR spectrum parameters caused by this process. High-speed (almost immediate) nanoparticle–cell interaction, resulting in a wide triplet (ΔH = 7.55 mT, 7.53 mT, 7.29 mT;
Figure 3d,f,h) presence in the ESR spectrum, was observed for all cells studied. According to the literature reports, endocytosis is divided depending on the process rate. Rapid endocytosis refers to the processes taking place in milliseconds, which distinguishes it from those much slower in time [
37]. Thus, in our opinion, the spectrum in
Figure 3d refers to the nanoparticles attached to a cell membrane. It results from very long correlation time at 240 K (τ = 10
−7 s) and no interaction between the nanoparticles (see
Figure 3b,d for comparison).
2.2. Effect of Incubation Time
Another important factor influencing nanoparticle–cell interaction is incubation time of cells with the nanoparticles. The examples of such temporal changes in ESR spectra are shown in
Figure 4.
To thoroughly examine the mechanisms responsible for changes in the structure of ESR spectra related to incubation time of nanoparticle solutions with cells, several experiments with different concentrations of nanoparticles/cells and different incubation times at 310 K were carried out.
Figure 4 shows how the ESR spectra of the nanoparticle solution with cells change over time in terms of their structure and intensity. An exemplary result of changes in the ESR spectrum intensity in time is presented in
Figure 5.
The intensity of the spin label (
Figure 4a–d and
Figure 5) attached to the magnetic nanoparticles decreased as the incubation time increased, and it reached a value close to zero after a few hours. The rate of signal loss depends on the nanoparticle concentration, the number of cells in the incubated sample, and the incubation conditions. On the other hand, one would suppose that the decrease of the ESR spectrum intensity could be caused by the reduced number of cells in a solution. Therefore, microscopic photos (
Figure 6) were taken to exclude such a reason for the observed changes. After analysis of a series of pictures taken, it was found that the number of cells increased with the incubation time. For example, the number of cells in yeast culture after 3 h of mixing them with the nanoparticles was approximately 2.4 times higher than initially while incubated without nanoparticles 2.6 (
Table 5). These values do not differ significantly. Additionally, this is another confirmation that the nanoparticles studied do not exhibit cytotoxicity on yeast cells. This clearly shows that the decrease in ESR signal intensity (
Figure 5) is due to the recombination of radicals and not to the decrease in cell number.
In the next step, the penetration of the functionalized nanoparticles into the interior of yeast cells was investigated. A series of confocal microscope images were taken (
Figure 7) using the fluorescein-labeled nanoparticles, Fe
3O
4@SiO
2@FITC@Dextran-TEMPOL. The pictures confirmed the presence of the nanoparticles inside yeast cells.
The incubation time influenced not only the ESR spectrum intensity but also its structure. Initially, the nanoparticles with the attached spin label interacted with cells (especially with a cell membrane) and the ESR spectrum recorded at 240 K was characterized by the presence of a wide triplet with a larger or smaller participation of a narrow triplet interpreted as the presence of the nanoparticles inside cells (
Figure 3d and
Figure 4a). As the incubation continued (at 310 K), the narrow triplet (
Figure 4b) gradually dominated until the ESR spectrum was fully converted to the narrow triplet (
Figure 4c). The rate of these changes strongly depends on the nanoparticle concentration, the number of cells, their type, and viability. The measurements taken at a temperature preventing endocytosis (274 K), using cooled solutions of the nanoparticles and cells, give ESR spectra with a wide triplet, and a narrow triplet does not appear over time. The ESR spectrum is similar, then, to that shown in
Figure 8a. Increasing the incubation temperature from 274 K to 310 K, when the endocytosis process was efficient, the narrow triplet was observed in the ESR spectrum (
Figure 8c).
On the other hand, if the sample is filtered (0.45 µm), no spin label signals are observed in the filtrate. However, it is visible in the material left on the filter. It proves that the narrow triplet comes from cell-associated spin labels, not from molecules in extracellular solution.
The dynamics and structure of the ESR spectra were changed during the experiments. Therefore, the EasySpin simulation program was used to determine several parameters, including spin label correlation times.
Figure 8 shows the results of the spectrum simulation with the so-called wide triplet (
Figure 8a,b) and narrow triplet (
Figure 8c,d). The simulations were carried out for the following spin label spectroscopic parameters: g factor (2.0027, 2.0069, 2. 0085), hyperfine splitting A (0.67, 0.65, 3.86 mT), and correlation time τ = 10
−7 s (
Figure 8b), τ = 10
−9 s (
Figure 8d).
2.3. Modeling of Endocytosis Process
The interaction process of the nanoparticles studied with cells can be divided into three main steps. The first one corresponds to the beginning of interacting nanoparticles–cells. In this early step, the ESR spectrum of a free radical 4-Hydroxy-TEMPO (TEMPOL) recorded at 240 K is characterized by a single line, similar in shape and parameters to that of the nanoparticle solution without cells (
Figure 3b and
Figure 9A). The second step is characterized by the differentiation of the ESR spectrum to a wide triplet (
Figure 8a and
Figure 9B) and extended correlation time (τ = 10
−7 s). The third (last) step is characterized by the presence of a narrow triplet (
Figure 8c and
Figure 9C) and a short correlation time (τ = 10
−9 s).
The entire process of changes in the structure and intensity of ESR spectrum during incubation of the nanoparticles with cells can be interpreted as follows. After mixing the nanoparticle solution with cells at a temperature that inhibits endocytosis (273 K), the magnetic core and the spin labels interact strongly with each other and do not bind to cells. It results in the presence of a single narrow line in the ESR spectrum recorded at 240 K (
Figure 9A). After a certain incubation period at 310 K, the nanoparticles attach to a cell—to be more precise, probably to a cell membrane. A wide triplet reflects this as an effect of longer correlation time and reduced interactions between spin labels (
Figure 8a and
Figure 9B). In the third step, the nanoparticles are located inside cells (as shown in
Figure 1F–G), presumably in organelles such as endosome/lysosome, however, further in vitro testing has to be done. It promotes rapid molecular movements and short correlation time at 240 K, and the spectrum is in the form of a narrow triplet (
Figure 8c and
Figure 9C). As the incubation time increases, the recombination of spin labels begins as a result of reactions with radicals, reactive oxygen species, and so forth. These reactions probably occur in cellular mitochondria. After several dozen minutes to several hours, the concentration of spin labels in a solution decreases several times (
Figure 9D), until there is a complete disappearance of spin label radicals. The period depends on the spin label type and coverage of the magnetic core.
Both the shape of the ESR spectrum (narrow triplet) and the short correlation time (τ = 10−9 s) testify to the high dynamics of movement of the TEMPOL spin label attached to magnetic nanoparticles. This situation is possible only inside the cell because at 240 K (ESR spectra recording temperature), the external environment is frozen. Thus, these results confirm the presence of nanoparticles in the interior of the cell and the usefulness of the ESR method for monitoring the process of endocytosis in cells.
Modeling of the endocytosis process of the nanoparticles having spin labels is very complex. We present here our attempt to model it. In the second step of the endocytosis process—penetration of nanoparticles into the cell—differences in nanoparticle concentration outside and inside cells can be approximated by Equations (1) and (2). The changes in concentration during incubation of cells with nanoparticles in the extracellular environment, A
ext, caused by entering of nanoparticles inside the cell will be described by Equation (1):
while those inside the cell, A
int, by Equation (2):
The total concentration of spin label should be constant over time in the absence of spin label recombination. According to experimental data (
Figure 5), the intensity of the ESR spin label signal decreases rapidly during incubation. It proves the participation of spin labels attached to the nanoparticles in redox reactions occurring most likely in cellular mitochondria. The whole process of ESR spectrum intensity changes can be approximated using Equation (3). The following assumptions were made: exponential decrease in intensity as a result of spin label radical recombination, a small correction on recombination rate during incubation as a result of a change in the number of cells in solution, and a decrease in recombination efficiency (
Figure 10).
where A
tot is total intensity of ESR spin label, b = 0.17, a = 500, c = 6 ∗ 10
−6 are fixed coefficients, and t is time.
Recombination of spin labels attached to a magnetite core proceeds by connecting protons to a spin label radical (
Figure 11). However, a detailed description and location of such reactions require further studies. The process probably occurs in cellular mitochondria as an effect of aerobic respiration, intense enzymatic responses, and generation of reactive oxygen species (ROS).