Antibody Conjugated PLGA Nanocarriers and Superparmagnetic Nanoparticles for Targeted Delivery of Oxaliplatin to Cells from Colorectal Carcinoma

Anti-CD133 monoclonal antibody (Ab)-conjugated poly(lactide-co-glycolide) (PLGA) nanocarriers, for the targeted delivery of oxaliplatin (OXA) and superparamagnetic nanoparticles (IO-OA) to colorectal cancer cells (CaCo-2), were designed, synthesized, characterized, and evaluated in this study. The co-encapsulation of OXA and IO-OA was achieved in two types of polymeric carriers, namely, PLGA and poly(lactide-co-glycolide)-poly(ethylene glycol) (PLGA-PEG) by double emulsion. PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA nanoparticles displayed a comparable mean diameter of 207 ± 70 nm and 185 ± 119 nm, respectively. The concentration of the released OXA from the PEGylated PLGA_IO-OA_OXA increased very rapidly, reaching ~100% release after only 2 h, while the PLGA_IO-OA_OXA displayed a slower and sustained drug release. Therefore, for a controlled OXA release, non-PEGylated PLGA nanoparticles were more convenient. Interestingly, preservation of the superparamagnetic behavior of the IO-OA, without magnetic hysteresis all along the dissolution process, was observed. The non-PEGylated nanoparticles (PLGA_OXA, PLGA_IO-OA_OXA) were selected for the anti-CD133 Ab conjugation. The affinity of Ab-coated nanoparticles for CD133-positive cells was examined using fluorescence microscopy in CaCo-2 cells, which was followed by a viability assay.


Introduction
According to the Global Cancer Observatory, colorectal cancer is the third most commonly diagnosed cancer, with ca. 1.93 million new cases diagnosed in 2020, making it the second major cause of cancer deaths [1,2]. This disease can be associated with socioeconomic development around the world, as changes in lifestyle and nutrition are key components on the impact on the intestinal microbiota associated with the risk of developing colorectal cancer [3,4]. The first-line treatment of colorectal cancer began in 1957, with chemotherapy based on 5-fluorouracil (5-FU) as an inhibitor of thymidylate synthase, which leads to the blocking of tumor cell division [5,6]. Later, further key progress was achieved, where the inhibition of thymidylate synthase could be potentiated by co-administration with reduced folate (e.g., leucovorin [5-formyl tetrahydrofolate]) as it increased the binding efficacy of 5-FU [7,8]. In the early 2000s, the co-administration of irinotecan (prodrug) and/or oxaliplatin (OXA) with 5-FU/leucovorin was associated with further enhanced cytotoxicity against metastatic colorectal cancer cells, thus preventing the DNA from unwinding and ultimately, cell death [9,10]. The next step was the development of treatments

Physicochemical Characteristics of the Iron Oxide Nanoparticles
The preparation of oleic acid-coated iron oxide nanoparticles (IO-OA), with a narrow size distribution, was achieved by the co-precipitation method using oleic acid (OA) as the coating agent and further dispersed in an organic phase (e.g., chloroform). According to the DLS measurements, IO-OA displayed a mean diameter of 20 ± 4 nm and zeta potential of −36 ± 4 mV ( Table 1). The negative zeta potential value indicates the coating of IO with OA, which confers them good colloidal stability and it is in agreement with our previous study [33]. XRD analysis confirmed the superparamagnetic structure of IO-OA nanoparticles, i.e., Fe 3 O 4 ( Figure S1), where X-ray peaks were identified similarly to previous work [33]. The superparamagnetic properties of the prepared nanoparticles were analyzed by VSM with a coercivity value of 3.91 Oe and a saturation magnetization (Ms) of 0.43 emu/g. The nanoparticles showed paramagnetism behavior without magnetic hysteresis at room temperature (25 • C), as depicted in Figure 1, which makes them act as good MRI contrast agents or suitable for treatment with magnetically mediated hyperthermia [38,44].

Chemical Structure of the Diblock Copolymer PLGA-PEG-NH 2
The poly(lactide-co-glycolide)-poly(ethylene glycol) bearing amino end group (PLGA-PEG-NH 2 ) diblock copolymer was synthesized as a step towards PEGylation of the PLGA nanoparticles to enhance their particle stability, prolong their circulation time when administered in the body and reduce their intermolecular aggregation through inter-particle steric repulsion [45][46][47]. Following the procedure reported in our previous work [33], the activation of the PLGA matrix was achieved by carbodiimide chemistry, using DCC/NHS and the subsequent attachment of the PEG diamine (NH 2 -PEG-NH 2 ). The conjugation of the PEG diamine to the PLGA backbone was revealed by 1 H-NMR, FTIR, and SEC. 1 H-NMR spectra, depicted in Figure 2A, attest to the covalent attachment of the PEG diamine onto PLGA end chains. The chemical shifts identified at 1.59 ppm, 3.61-3.64 ppm, 4.50-4.90 ppm, and 5. 13-5.30 ppm correspond to the -CH 3 of the lactide, -CH 2 -of the ethylene glycol, -CH 2 -of the glycolide, and -CH-of the lactide, respectively. FTIR analysis further confirmed the formation of the diblock copolymer ( Figure 2B), where the absorption band at 2883 cm −1 ascribed to the C-H stretching vibration of the methylene group of PEG chains was noticed in the PLGA-PEG-NH 2 spectrum, confirming the reproducibility of the synthesis and in agreement with our reported work [33]. * Polydispersity index (PDI) with standard deviation (from 3 replicates).

Chemical Structure of the Diblock Copolymer PLGA-PEG-NH2
The poly(lactide-co-glycolide)-poly(ethylene glycol) bearing amino end group (PLGA-PEG-NH2) diblock copolymer was synthesized as a step towards PEGylation of the PLGA nanoparticles to enhance their particle stability, prolong their circulation time when administered in the body and reduce their intermolecular aggregation through inter-particle steric repulsion [45][46][47]. Following the procedure reported in our previous work [33], the activation of the PLGA matrix was achieved by carbodiimide chemistry, using DCC/NHS and the subsequent attachment of the PEG diamine (NH2-PEG-NH2). The conjugation of the PEG diamine to the PLGA backbone was revealed by 1 H-NMR, FTIR, and SEC. 1 H-NMR spectra, depicted in Figure 2A, attest to the covalent attachment of the PEG diamine onto PLGA end chains. The chemical shifts identified at 1.59 ppm, 3.61-3.64 ppm, 4.50-4.90 ppm, and 5.13-5.30 ppm correspond to the -CH3 of the lactide, -CH2-of the ethylene glycol, -CH2-of the glycolide, and -CH-of the lactide, respectively. FTIR analysis further confirmed the formation of the diblock copolymer ( Figure 2B), −1 Figure 1. Magnetic properties at room temperature (25 • C) of oleic acid-coated iron oxide nanoparticles (IO-OA) synthesized by co-precipitation. The measurement was performed in the magnetic field from −10 kOe to +10 kOe.

Multicomponent Drug Delivery Systems: Mean Size, Morphology, and Surface Properties
The multicomponent drug delivery systems, namely, PLGA_IO-OA_OXA, and PEGylated PLGA_IO-OA_OXA, were produced by the double emulsion method [42]. As a frame of reference, polymeric nanoparticles loaded with a single therapeutic agent were also prepared, namely, PLGA_IO-OA, PLGA_OXA, PEGylated PLGA_IO-OA, and PEGylated PLGA_OXA. methylene group of PEG chains was noticed in the PLGA-PEG-NH2 spectrum, confirming the reproducibility of the synthesis and in agreement with our reported work [33].

Multicomponent Drug Delivery Systems: Mean Size, Morphology, and Surface Properties
The multicomponent drug delivery systems, namely, PLGA_IO-OA_OXA, and PEGylated PLGA_IO-OA_OXA, were produced by the double emulsion method [42]. As a frame of reference, polymeric nanoparticles loaded with a single therapeutic agent were also prepared, namely, PLGA_IO-OA, PLGA_OXA, PEGylated PLGA_IO-OA, and PEGylated PLGA_OXA.
The size distribution of all the nanoparticles measured by DLS is depicted in Figure  S2 and the average values are summarized in Table 1. Based on the DLS measurements, all produced nanoparticles, i.e., PLGA, PLGA_OXA, PLGA_IO-OA, PLGA_IO-OA_OXA, PEGylated PLGA, PEGylated PLGA_OXA, PEGylated PLGA_IO-OA, and PEGylated PLGA_IO-OA_OXA exhibited a comparable mean diameter within the standard deviation. For instance, PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA displayed a mean diameter of 207 ± 70 nm and 185 ± 119 nm, respectively. These results show that the encapsulation of OXA and/or IO-OA and the PEGylation process do not affect, statistically, the hydrodynamic size of the nanoparticles (p ≥ 0.05) [48]. The mean size of nanoparticles is usually affected rather by the sonication power and time, and by the amount of the surfactant [30]. Conversely, the zeta potential values varied between the PEGylated and non-PEGylated analogs, e.g., −8 ± 7 mV for PLGA_IO-OA_OXA and 4 ± 3 mV for PEGylated PLGA_IO-OA_OXA. This discrepancy in the zeta potential values is associated with the nature of the functional groups present on the surface of the nanoparticles. The inversion of the zeta potential values from negative in PLGA-based nanoparticles to positive in the PEGylated analogs is consistent with the presence of amine groups in the latter from the PLGA-PEG diblock copolymer. It should be noted that the zeta potential values of the nanoparticles remained the same within the standard deviation in the PBS buffer. Examples of TEM images of the prepared nanoparticles are depicted in Figure 3. All of them exhibited a spherical shape in size ranges of hundreds of nanometers. Image treatment of TEM pictures using ImageJ software indicated a smaller average diameter (Table  S1) of the nanoparticles compared to the DLS measurements. This discrepancy is associated with the principle of analysis of each method. Indeed, DLS measures the mean hydrodynamic diameter of nanoparticles dispersed in an aqueous solution, while TEM analysis allows us to determine the average diameter of dehydrated nanoparticles. They also showed the encapsulation of the IO-OA within the polymeric nanoparticles ( Figure  3C,D,G,H). It is worth noting that the high PDI presented in PEG-conjugated nanoparticles is associated with their slight lack of stability, compared to their non PEG-conjugated The size distribution of all the nanoparticles measured by DLS is depicted in Figure S2 and the average values are summarized in Table 1. Based on the DLS measurements, all produced nanoparticles, i.e., PLGA, PLGA_OXA, PLGA_IO-OA, PLGA_IO-OA_OXA, PEGylated PLGA, PEGylated PLGA_OXA, PEGylated PLGA_IO-OA, and PEGylated PLGA_IO-OA_OXA exhibited a comparable mean diameter within the standard deviation. For instance, PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA displayed a mean diameter of 207 ± 70 nm and 185 ± 119 nm, respectively. These results show that the encapsulation of OXA and/or IO-OA and the PEGylation process do not affect, statistically, the hydrodynamic size of the nanoparticles (p ≥ 0.05) [48]. The mean size of nanoparticles is usually affected rather by the sonication power and time, and by the amount of the surfactant [30]. Conversely, the zeta potential values varied between the PEGylated and non-PEGylated analogs, e.g., −8 ± 7 mV for PLGA_IO-OA_OXA and 4 ± 3 mV for PEGylated PLGA_IO-OA_OXA. This discrepancy in the zeta potential values is associated with the nature of the functional groups present on the surface of the nanoparticles. The inversion of the zeta potential values from negative in PLGA-based nanoparticles to positive in the PEGylated analogs is consistent with the presence of amine groups in the latter from the PLGA-PEG diblock copolymer. It should be noted that the zeta potential values of the nanoparticles remained the same within the standard deviation in the PBS buffer. Examples of TEM images of the prepared nanoparticles are depicted in Figure 3. All of them exhibited a spherical shape in size ranges of hundreds of nanometers. Image treatment of TEM pictures using ImageJ software indicated a smaller average diameter (Table S1) of the nanoparticles compared to the DLS measurements. This discrepancy is associated with the principle of analysis of each method. Indeed, DLS measures the mean hydrodynamic diameter of nanoparticles dispersed in an aqueous solution, while TEM analysis allows us to determine the average diameter of dehydrated nanoparticles. They also showed the encapsulation of the IO-OA within the polymeric nanoparticles ( Figure 3C,D,G,H). It is worth noting that the high PDI presented in PEG-conjugated nanoparticles is associated with their slight lack of stability, compared to their non PEG-conjugated analogs. Indeed, the presence of PEG on the surface of PLGA may lead to a coalescence-like behavior between some individual nanoparticles, due to the low zeta-potential (i.e., between 0 and 5 mV). This behavior can be also seen in the TEM images ( Figure 3G,H).
The colloidal stability of the nanoparticles was monitored over 15 days, as can be seen in Figure 4. The mean diameter, as well as the zeta potential values for all nanoparticles, including the multicomponent drug delivery systems, was maintained without any significant changes (p ≥ 0.05).
analogs. Indeed, the presence of PEG on the surface of PLGA may lead to a coalescencelike behavior between some individual nanoparticles, due to the low zeta-potential (i.e., between 0 and 5 mV). This behavior can be also seen in the TEM images ( Figure 3G,H).
The colloidal stability of the nanoparticles was monitored over 15 days, as can be seen in Figure 4. The mean diameter, as well as the zeta potential values for all nanoparticles, including the multicomponent drug delivery systems, was maintained without any significant changes (p ≥ 0.05).

Drug Loading and Drug Release Kinetics
The determination of the content of OXA in the drug delivery nanoparticles is crucial for defining the drug loading and encapsulation efficiency. The UV-visible spectrophotometer failed in determining the drug loading with high accuracy, since OXA absorbs in the same UV region as the PLGA polymer, in the range of 180-217 nm ( Figure S3). The use of multiple extraction steps to isolate OXA from the polymer with high extraction

Drug Loading and Drug Release Kinetics
The determination of the content of OXA in the drug delivery nanoparticles is crucial for defining the drug loading and encapsulation efficiency. The UV-visible spectrophotometer failed in determining the drug loading with high accuracy, since OXA absorbs in the same UV region as the PLGA polymer, in the range of 180-217 nm ( Figure S3). The use of multiple extraction steps to isolate OXA from the polymer with high extraction efficiency was unsuccessful. To overcome this issue, TGA analysis was used to quantify the OXA encapsulated in the nanoparticles ( Figure 5), since this drug possesses Pt in its chemical structure, which remains undegradable during the thermal cycle up to 800 • C (the highest applied temperature). According to the TG curve of pure OXA, the total weight loss corresponding to the decomposition of the organic part reaches 50% at 800 • C and, hence, the remaining 50% of the mass corresponds to Pt [49]. As the total weight loss of the PLGA and PLGA-PEG-NH 2 matrices is 100% at 800 • C, hence the TG curves of PLGA_OXA and PEGylated PLGA_OXA nanoparticles, the remaining mass of 2% at 800 • C corresponds to the Pt present in the nanoparticles, which represents half of the total weight fraction of OXA in the nanoparticles. The total mass of OXA encapsulated in the nanoparticles corresponds to 4%. Based on the TGA results, a drug loading of 22% and an encapsulation efficiency of 44% were calculated in both PLGA_OXA and PEGylated PLGA_OXA. The PEGylation of the PLGA nanoparticles does not have an impact on the encapsulation efficiency, which is in good agreement with our previous study using a different drug [33]. TGA analysis cannot be used to determine the drug loading of OXA in the multicomponent PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA, since both contain IO, as Pt does not decompose at the maximum applied temperature (800 • C). Nevertheless, our previous study clearly showed, with very good reproducibility of the experiments, that the encapsulation of an additional therapeutic agent (i.e., IO) besides the drug, within a PLGA/PEGylated PLGA nanocarrier, does not affect the encapsulation efficiency [33]. Based on these findings, it is reasonable to conclude that the drug loading and encapsulation efficiencies in PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA nanoparticles were the same as for PLGA_OXA and PEGylated PLGA_OXA nanoparticles. efficiency was unsuccessful. To overcome this issue, TGA analysis was used to quantify the OXA encapsulated in the nanoparticles ( Figure 5), since this drug possesses Pt in its chemical structure, which remains undegradable during the thermal cycle up to 800 °C (the highest applied temperature). According to the TG curve of pure OXA, the total weight loss corresponding to the decomposition of the organic part reaches 50% at 800 °C and, hence, the remaining 50% of the mass corresponds to Pt [49]. As the total weight loss of the PLGA and PLGA-PEG-NH2 matrices is 100% at 800 °C, hence the TG curves of PLGA_OXA and PEGylated PLGA_OXA nanoparticles, the remaining mass of 2% at 800 °C corresponds to the Pt present in the nanoparticles, which represents half of the total weight fraction of OXA in the nanoparticles. The total mass of OXA encapsulated in the nanoparticles corresponds to 4%. Based on the TGA results, a drug loading of 22% and an encapsulation efficiency of 44% were calculated in both PLGA_OXA and PEGylated PLGA_OXA. The PEGylation of the PLGA nanoparticles does not have an impact on the encapsulation efficiency, which is in good agreement with our previous study using a different drug [33]. TGA analysis cannot be used to determine the drug loading of OXA in the multicomponent PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA, since both contain IO, as Pt does not decompose at the maximum applied temperature (800 °C). Nevertheless, our previous study clearly showed, with very good reproducibility of the experiments, that the encapsulation of an additional therapeutic agent (i.e., IO) besides the drug, within a PLGA/PEGylated PLGA nanocarrier, does not affect the encapsulation efficiency [33]. Based on these findings, it is reasonable to conclude that the drug loading and encapsulation efficiencies in PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA nanoparticles were the same as for PLGA_OXA and PEGylated PLGA_OXA nanoparticles. Prior to Ab conjugation, the drug release profiles of the multicomponent drug delivery systems PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA were examined and compared to those of their drug-loaded nanoparticle analogs free of IO-OA, namely, PLGA_OXA and PEGylated PLGA_OXA. The dissolution tests were monitored over several days in PBS at 37 °C ( Figure 6 and Table S2). The concentration of the released OXA from the PEGylated PLGA_IO-OA_OXA increased very rapidly during the first minutes, Prior to Ab conjugation, the drug release profiles of the multicomponent drug delivery systems PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA were examined and compared to those of their drug-loaded nanoparticle analogs free of IO-OA, namely, PLGA_OXA and PEGylated PLGA_OXA. The dissolution tests were monitored over several days in PBS at 37 • C ( Figure 6 and Table S2). The concentration of the released OXA from the PEGylated PLGA_IO-OA_OXA increased very rapidly during the first minutes, reaching~100% release (40 ± 4 µg·mL −1 ) after only 2 h. In PEGylated PLGA_ OXA free of IO-OA, the drug release profile was characterized by an initial burst release of~75%, which occurred during the first 2 h, followed by a slower and controlled drug release, until reaching~100% release (43 ± 8 µg·mL −1 ) after 96 h. The super-fast and non-controlled OXA release in PEGylated PLGA_IO-OA_OXA, taking place during the burst release phase, is probably associated with the hydrophilic nature of OXA, the PEGylation of the nanoparticles, and the presence of IO-OA [33,34]. Owing to their hydrophilic nature, OXA may preferentially locate close to or within the PEG shell. This phenomenon might be enhanced in the presence of another hydrophobic therapeutic agent, such as IO-OA, in the nanoparticles, as they can induce more favorable interactions than OXA with the PLGA matrix. Conversely, PLGA_OXA and PLGA_IO-OA_OXA displayed a slower and more sustained drug release compared to their PEGylated analogs. The dissolution profile of PLGA_IO-OA_OXA exhibited a prominent initial burst release of about 45% during the first 2 h, which was followed by a slower release, until reaching~100% release (40 ± 4 µg·mL −1 ) after 168 h. Interestingly, PLGA_OXA free of IO-OA displayed a triphasic profile, characterized by an initial burst release of about 20% during the first 2 h. This is followed by diffusion, swelling, and lastly an accelerated drug release phase, caused by the polymer degradation, attaining~100% release (45 ± 4 µg·mL −1 ) after 264 h [50]. The faster drug release observed in the presence of IO-OA (i.e., PLGA_IO-OA_OXA) is caused by the catalytic activity of the magnetic nanoparticles, leading to faster degradation of the PLGA matrix. Moreover, the faster drug release, noticed in the PEGylated nanoparticles compared to their non-PEGylated analogs, is due to the hydrophilic nature of PEG chains that thereby facilitate the water uptake and accelerate the degradation of the PLGA matrix by hydrolysis. The effect of PEGylation on the degradation kinetics of PLGA during the drug release process has already been described in the literature [33,51]. Therefore, for a controlled OXA release, non-PEGylated PLGA multicomponent systems are more suitable. Based on these findings, Ab conjugation was carried out on PLGA_IO-OA_OXA.
reaching ~100% release (40 ± 4 µg·mL −1 ) after only 2 h. In PEGylated PLGA_ OXA free of IO-OA, the drug release profile was characterized by an initial burst release of ~75%, which occurred during the first 2 h, followed by a slower and controlled drug release, until reaching ~100% release (43 ± 8 µg·mL −1 ) after 96 h. The super-fast and non-controlled OXA release in PEGylated PLGA_IO-OA_OXA, taking place during the burst release phase, is probably associated with the hydrophilic nature of OXA, the PEGylation of the nanoparticles, and the presence of IO-OA [33,34]. Owing to their hydrophilic nature, OXA may preferentially locate close to or within the PEG shell. This phenomenon might be enhanced in the presence of another hydrophobic therapeutic agent, such as IO-OA, in the nanoparticles, as they can induce more favorable interactions than OXA with the PLGA matrix. Conversely, PLGA_OXA and PLGA_IO-OA_OXA displayed a slower and more sustained drug release compared to their PEGylated analogs. The dissolution profile of PLGA_IO-OA_OXA exhibited a prominent initial burst release of about 45% during the first 2 h, which was followed by a slower release, until reaching ~100% release (40 ± 4 µg·mL −1 ) after 168 h. Interestingly, PLGA_OXA free of IO-OA displayed a triphasic profile, characterized by an initial burst release of about 20% during the first 2 h. This is followed by diffusion, swelling, and lastly an accelerated drug release phase, caused by the polymer degradation, attaining ~100% release (45 ± 4 µg·mL −1 ) after 264 h [50]. The faster drug release observed in the presence of IO-OA (i.e., PLGA_IO-OA_OXA) is caused by the catalytic activity of the magnetic nanoparticles, leading to faster degradation of the PLGA matrix. Moreover, the faster drug release, noticed in the PEGylated nanoparticles compared to their non-PEGylated analogs, is due to the hydrophilic nature of PEG chains that thereby facilitate the water uptake and accelerate the degradation of the PLGA matrix by hydrolysis. The effect of PEGylation on the degradation kinetics of PLGA during the drug release process has already been described in the literature [33,51]. Therefore, for a controlled OXA release, non-PEGylated PLGA multicomponent systems are more suitable. Based on these findings, Ab conjugation was carried out on PLGA_IO-OA_OXA.  Besides the drug release performance of the multicomponent drug delivery systems, i.e., PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA, their magnetic properties were examined before and after dissolution tests (Figure 7). The magnetization curves of both PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA present zero remanence and zero coercivity before, as well as after, the dissolution tests. The saturation magnetization for PLGA_IO-OA_OXA reached a value of 0.0019 emu·g −1 with a coercivity of 2.89 Oe, and at the end dissolution test, this value increased to 0.0091 emu·g −1 with a coercivity of 3.81 Oe. Likewise, the same trend was observed in the case of PEGylated PLGA_IO-OA_OXA, for which an increase in the saturation magnetization value was detected, from 0.0021 emu·g −1 (coercivity of 3.07 Oe) before dissolution tests to 0.0051 emu·g −1 (coercivity of 3.61 Oe) after 96 h of drug release (the end of dissolution test). The increase in both saturation magnetization and coercivity values after complete drug release can be explained by the degradation of the polymeric matrix, leading to the apparent increase in the concentration of IO-OA in the remaining solid content. The preservation of the superparamagnetic behavior of the IO-OA, without magnetic hysteresis all along the dissolution process, makes the multicomponent drug delivery systems suitable for treatment with magnetically mediated hyperthermia, even after the drug release is completed.
both PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA present zero remanence and zero coercivity before, as well as after, the dissolution tests. The saturation magnetization for PLGA_IO-OA_OXA reached a value of 0.0019 emu·g −1 with a coercivity of 2.89 Oe, and at the end dissolution test, this value increased to 0.0091 emu·g −1 with a coercivity of 3.81 Oe. Likewise, the same trend was observed in the case of PEGylated PLGA_IO-OA_OXA, for which an increase in the saturation magnetization value was detected, from 0.0021 emu·g −1 (coercivity of 3.07 Oe) before dissolution tests to 0.0051 emu·g −1 (coercivity of 3.61 Oe) after 96 h of drug release (the end of dissolution test). The increase in both saturation magnetization and coercivity values after complete drug release can be explained by the degradation of the polymeric matrix, leading to the apparent increase in the concentration of IO-OA in the remaining solid content. The preservation of the superparamagnetic behavior of the IO-OA, without magnetic hysteresis all along the dissolution process, makes the multicomponent drug delivery systems suitable for treatment with magnetically mediated hyperthermia, even after the drug release is completed.

In Vitro Antitumor Activity of PLGA_IO-OA_OXA_Ab
To develop a multicomponent and multifunctional delivery system, capable of both drug delivery and cellular targeting, Ab-conjugated PLGA nanoparticles were prepared by covalent attachment of the anti-CD133 Ab to the surface of PLGA_IO-OA_OXA, further labeled as PLGA_IO-OXA_Ab. For comparison, PLGA_OXA_Ab nanoparticles were also prepared. As can be seen in Table 1, after Ab conjugation, the mean diameter, as well as the zeta potential of PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab, were comparable to their unconjugated analogs when taking into account the standard deviation. The coating of the nanoparticle surface by Ab is apparent in the TEM pictures in Figure 8B,C.

In Vitro Antitumor Activity of PLGA_IO-OA_OXA_Ab
To develop a multicomponent and multifunctional delivery system, capable of both drug delivery and cellular targeting, Ab-conjugated PLGA nanoparticles were prepared by covalent attachment of the anti-CD133 Ab to the surface of PLGA_IO-OA_OXA, further labeled as PLGA_IO-OXA_Ab. For comparison, PLGA_OXA_Ab nanoparticles were also prepared. As can be seen in Table 1, after Ab conjugation, the mean diameter, as well as the zeta potential of PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab, were comparable to their unconjugated analogs when taking into account the standard deviation. The coating of the nanoparticle surface by Ab is apparent in the TEM pictures in Figure 8B,C.
Regarding in vitro experiments in human cells, first, we aimed to evaluate the binding specificity of PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab in CaCo-2 cells, which should, according to the literature, express the CD133 marker, the RNA levels of which have been documented to be pronounced in CaCo-2 cells in comparison to other cell lines [52]. Therefore, to verify that the CaCo-2 cells held their characteristics, remained unchanged, and produced the CD133 expression also on the protein level, fluorescently labeled Ab anti-CD133 binding assay was performed in these cells. The staining of this surface protein was positive, which is documented in Figure 9, with fluorescence emission detected on the cell surface, meaning that CD133 is expressed at the cellular membrane of CaCo-2 cells. Regarding in vitro experiments in human cells, first, we aimed to evaluate the binding specificity of PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab in CaCo-2 cells, which should, according to the literature, express the CD133 marker, the RNA levels of which have been documented to be pronounced in CaCo-2 cells in comparison to other cell lines [52]. Therefore, to verify that the CaCo-2 cells held their characteristics, remained unchanged, and produced the CD133 expression also on the protein level, fluorescently labeled Ab anti-CD133 binding assay was performed in these cells. The staining of this surface protein was positive, which is documented in Figure 9, with fluorescence emission detected on the cell surface, meaning that CD133 is expressed at the cellular membrane of CaCo-2 cells.   Regarding in vitro experiments in human cells, first, we aimed to evaluate the binding specificity of PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab in CaCo-2 cells, which should, according to the literature, express the CD133 marker, the RNA levels of which have been documented to be pronounced in CaCo-2 cells in comparison to other cell lines [52]. Therefore, to verify that the CaCo-2 cells held their characteristics, remained unchanged, and produced the CD133 expression also on the protein level, fluorescently labeled Ab anti-CD133 binding assay was performed in these cells. The staining of this surface protein was positive, which is documented in Figure 9, with fluorescence emission detected on the cell surface, meaning that CD133 is expressed at the cellular membrane of CaCo-2 cells.  Next, we aimed to investigate the localization and intracellular internalization of PLGA_OXA nanoparticles coated with anti-CD133 Ab (PLGA_OXA_Ab), which was done by binding study in living CaCo-2 cells, again using fluorescence microscopy, since this method provides a direct and facile approach for monitoring the intracellular trafficking and fate of the nanoparticles. The PLGA_OXA_Ab cell uptake was visualized by Alexa Fluor 488 (green emission), i.e., the presence of nanoparticles in CaCo-2 cells was detected using the FITC channel. Since, in general, nanoparticle uptake by cells is dosedependent, a range of nanoparticle concentrations for the localization study was used (specifically, 1.8-90 µg·mL −1 ). As is apparent from Figure 10 (images are false-colored based on the fluorescence emission intensity), even a concentration of the nanoparticles as low as 1.8 µg·mL −1 was sufficient for visualization of the CD133 protein binding to the cell surface. The images were taken 30 min after the nanoparticle treatment of the CaCo-2 cells and, as can be seen, some proportion of the NPs were already internalized in the cells; therefore, except for the cell membrane, the highest fluorescence emission intensity was apparent around the cell nucleus. This effect was likely caused by the specific antigen-antibody interaction/binding, with subsequent nanoparticle internalization likely by receptor-mediated endocytosis. Therefore, here, we verified that the anti-CD133 Ab can be used for targeting CD133-positive cells.
ing and fate of the nanoparticles. The PLGA_OXA_Ab cell uptake was visualized by Alexa Fluor 488 (green emission), i.e., the presence of nanoparticles in CaCo-2 cells was detected using the FITC channel. Since, in general, nanoparticle uptake by cells is dose-dependent, a range of nanoparticle concentrations for the localization study was used (specifically, 1.8-90 µg·mL −1 ). As is apparent from Figure 10 (images are false-colored based on the fluorescence emission intensity), even a concentration of the nanoparticles as low as 1.8 µg·mL −1 was sufficient for visualization of the CD133 protein binding to the cell surface. The images were taken 30 min after the nanoparticle treatment of the CaCo-2 cells and, as can be seen, some proportion of the NPs were already internalized in the cells; therefore, except for the cell membrane, the highest fluorescence emission intensity was apparent around the cell nucleus. This effect was likely caused by the specific antigen-antibody interaction/binding, with subsequent nanoparticle internalization likely by receptor-mediated endocytosis. Therefore, here, we verified that the anti-CD133 Ab can be used for targeting CD133-positive cells. The same data as for PLGA_OXA_Ab were gained for PLGA_IO-OA_OXA_Ab (see Figure S4 in the Supplementary Materials). This again, confirms the efficient cell surface binding of the nanoparticles and partial internalization, after only 30 min incubation. No differences were observed for nanoparticles containing iron oxide.
However, in contrast to the cancerous CaCo-2 cells, in noncancerous human primary fibroblasts (MRC-5), we have proven no PLGA_OXA_Ab binding ( Figure S5 in the Supplementary Materials), as expected since MRC-5 cells should not express the CD133 protein.
Next, to verify that upon PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab treatment, the detected fluorescence emission on the CaCo-2 cell membrane was caused by specific Ab binding to the CD133 antigen, a colocalization study using fluorescence microscopy was performed. In this experiment, CaCo-2 cells were treated with PLGA_OXA_Ab (90 µg·mL −1 ) for 30 min (green fluorescence emission) and co-stained with anti-CD133 Ab, conjugated to Atto 565 (red fluorescence emission). Based on the images in Figure 11, the co-localization proved to be successful. See the Supplementary Materials ( Figure S6) for a zoomed section of the PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab co-localization. Figure S4 in the Supplementary Materials). This again, confirms the efficient cell surface binding of the nanoparticles and partial internalization, after only 30 min incubation. No differences were observed for nanoparticles containing iron oxide.
However, in contrast to the cancerous CaCo-2 cells, in noncancerous human primary fibroblasts (MRC-5), we have proven no PLGA_OXA_Ab binding ( Figure S5 in the Supplementary Materials), as expected since MRC-5 cells should not express the CD133 protein.
Next, to verify that upon PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab treatment, the detected fluorescence emission on the CaCo-2 cell membrane was caused by specific Ab binding to the CD133 antigen, a colocalization study using fluorescence microscopy was performed. In this experiment, CaCo-2 cells were treated with PLGA_OXA_Ab (90 µg·mL −1 ) for 30 min (green fluorescence emission) and co-stained with anti-CD133 Ab, conjugated to Atto 565 (red fluorescence emission). Based on the images in Figure 11, the co-localization proved to be successful. See the Supplementary Materials ( Figure S6) for a zoomed section of the PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab co-localization. As a next step, we evaluated the effect of the OXA-containing nanoparticles on the cell metabolic activity of cancerous CaCo-2 and noncancerous MRC-5 cells, using WST-1 assay and 72 h incubation. Cell treatment with the PLGA_OXA showed a dose-response relationship. The results, summarized in Table 2, show that the IC50 corresponded to 255 ± 11 µg·mL −1 , in CaCo-2 cells expressing the CD133 receptor and to 525 ± 12 µg·mL −1 , in CD133-negative noncancerous MRC-5 cells (Table 3, expressed for the OXA content only). Pure nanoparticles (PLGA_blank), not containing OXA, did not exhibit cytotoxicity and did not reach IC50, up to the highest evaluated concentration of 900 µg·mL −1 . Finally, the number of living CaCo-2 cells was evaluated using calcein, after PLGA_OXA treatment ( Figure 12) for 72 h. There was a slight decrease in the proportion of living cells already from the 225 µg·mL −1 concentration. A drop to ca. one-third of living cells, in comparison As a next step, we evaluated the effect of the OXA-containing nanoparticles on the cell metabolic activity of cancerous CaCo-2 and noncancerous MRC-5 cells, using WST-1 assay and 72 h incubation. Cell treatment with the PLGA_OXA showed a dose-response relationship. The results, summarized in Table 2, show that the IC 50 corresponded to 255 ± 11 µg·mL −1 , in CaCo-2 cells expressing the CD133 receptor and to 525 ± 12 µg·mL −1 , in CD133-negative noncancerous MRC-5 cells (Table 3, expressed for the OXA content only). Pure nanoparticles (PLGA_blank), not containing OXA, did not exhibit cytotoxicity and did not reach IC 50 , up to the highest evaluated concentration of 900 µg·mL −1 . Finally, the number of living CaCo-2 cells was evaluated using calcein, after PLGA_OXA treatment ( Figure 12) for 72 h. There was a slight decrease in the proportion of living cells already from the 225 µg·mL −1 concentration. A drop to ca. one-third of living cells, in comparison to the untreated control, was detected at the concentration of 900 µg·mL −1 . As for the blank NPs (nanoparticles not containing OXA), no marked drop (in the range of SD) in the proportion of living cells was detected ( Figure S6 in the Supplementary Materials). Therefore, these findings suggest that PLGA nanoparticles themselves are nontoxic and suitable as nanocarriers for the delivery of various drugs, as well as being suitable for Ab conjugation and, thus, specific cell targeting. to the untreated control, was detected at the concentration of 900 µg·mL −1 . As for the blank NPs (nanoparticles not containing OXA), no marked drop (in the range of SD) in the proportion of living cells was detected ( Figure S6 in the Supplementary Materials). Therefore, these findings suggest that PLGA nanoparticles themselves are nontoxic and suitable as nanocarriers for the delivery of various drugs, as well as being suitable for Ab conjugation and, thus, specific cell targeting.

Discussion
Oxaliplatin (OXA) is considered one of the essential cytotoxic drugs for modern chemotherapy in colorectal cancer, as it has led to improvements in patients' survival and management of the disease [53]. Other platinum derivatives, such as cisplatin and carboplatin, possess intrinsic resistance that limits their application as therapeutic agents for colorectal cancer [54,55]. The chemical structure of OXA differs from cisplatin or carboplatin, as it possesses an oxalate and a diaminocyclohexane (DACH) group (supporting information, Figure S7). Part of the cytotoxic activity of OXA has been attributed to the DACH ligand, which is retained by the activated oxaliplatin, resulting in the inhibition of the DNA synthesis by the formation of interstrand crosslink and intrastrand platinum-DNA adducts, and it appears to be more effective and cytotoxic than adducts formed from

Discussion
Oxaliplatin (OXA) is considered one of the essential cytotoxic drugs for modern chemotherapy in colorectal cancer, as it has led to improvements in patients' survival and management of the disease [53]. Other platinum derivatives, such as cisplatin and carboplatin, possess intrinsic resistance that limits their application as therapeutic agents for colorectal cancer [54,55]. The chemical structure of OXA differs from cisplatin or carboplatin, as it possesses an oxalate and a diaminocyclohexane (DACH) group (Supplementary Materials, Figure S7). Part of the cytotoxic activity of OXA has been attributed to the DACH ligand, which is retained by the activated oxaliplatin, resulting in the inhibition of the DNA synthesis by the formation of interstrand crosslink and intrastrand platinum-DNA adducts, and it appears to be more effective and cytotoxic than adducts formed from cisplatin or carboplatin [56,57]. Even though OXA has demonstrated significant activity in clinical practice against colorectal cancer, it can induce peripheral neurotoxicity (IPN) in the form of acute and chronic clinical syndrome; this is still considered the major dose-limiting factor [21]. For patients that receive a cumulative dose of OXA, it is expected that high-grade chronic IPN will occur in about 10% of patients, from receiving doses between 510-766 mg·m −2 , while at doses higher than 1000 mg·m −2 , the likelihood of developing IPN can reach almost 50% of treated patients [58]. Therefore, to reduce OXA IPN, its encapsulation into polymeric nanoparticles comes as a promising approach for cancer treatment. For example, in a study by Tummala et al., the encapsulation of OXA was achieved in a chitosan-based pHresponsive hybrid as a therapeutic agent [59]. In another study, Handali et al. encapsulated 5-FU and OXA into biodegradable PHBV/PLGA nanoparticles, which exhibited a higher cytotoxicity effect than free drugs on cancer cells [60,61]. Encapsulation of therapeutic drugs using biodegradable polymers, such as PLGA, has attracted attention due to its biocompatibility, non-toxicity, and non-immunogenicity [62].
In this study, the design of the drug delivery system was not only aimed at encapsulating OXA in PLGA based nanoparticles, but at combining multiple functionalities into a single system. A combination of OXA with magnetic hyperthermia, through superparamagnetic iron oxide nanoparticles (IO), could improve diagnosis and cancer therapy, as magnetic hyperthermia enables one to target the heating source at the tumor tissues, through the application of an external magnetic field [63]. Promising results have been reported using multifunctional platforms for colorectal cancer therapy [64,65]. In a study by Lima et al., the platform was designed based on polyethylene glycol-polylactic acid (PEG-PLA) nanospheres to co-deliver methotrexate and IO nanoparticles, proving increased cytotoxicity towards Caco-2 and SW-480 colon cancer cells [66]. As tumor biology is also better understood, so is the possibility to apply a targeted multifunctional drug delivery system, as this will be essential for efficient use of the resources and better outcomes for patients.
For this study, the incorporation of OXA and IO nanoparticles into PLGA-based nanoparticles with a surface targeting agent (anti-CD133 Ab) for the treatment of colorectal cancer was designed. The assembly of the multicomponent drug delivery systems was achieved by the encapsulation of oleic acid-coated iron oxide nanoparticles (IO-OA) and/or OXA into a PLGA or PLGA-PEG-NH 2 polymer, using the double emulsion method. The reason that a PEGylated PLGA multicomponent system was designed is that the addition of PEG on the surface of the PLGA nanoparticles can give the targeted multicomponent drug delivery system a stealth barrier, which can delay opsonization and recognition by the reticuloendothelial system [67]. In vitro release investigations showed that the OXA from the multicomponent drug delivery systems, PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA, were faster for the PEGylated analog. The concentration of the released OXA from the PEGylated PLGA_IO-OA_OXA increased very rapidly, reaching~100% release after only 2 h, while the PLGA_IO-OA_OXA displayed a slower and sustained drug release. The super-fast and non-controlled OXA release in the PEGylated PLGA_IO-OA_OXA is probably connected with the hydrophilic nature of OXA, the PEGylation of the nanoparticles, and the presence of IO-OA. A similar observation of a quicker release, using a hydrophilic carrier (PEG-PLGA), was observed in the work of Babos et al., in which a water-soluble drug (doxorubicin HCl) was encapsulated into both PLGA-based and PEG-PLGA nanoparticles [68]. Therefore, for a targeted system with controlled release, the PLGA_IO-OA_OXA was chosen for the conjugation with the anti-CD133 antibody, conjugated to Alexa Fluor 488.
Based on the expression profile of CD133 in CaCo-2 cells [52], we utilized the interaction of anti-CD133 Ab-coated nanoparticles and CD133 cell surface antigens as a very promising approach to realize the highly specific and effective delivery of OXA to CD133overexpressing cells from colorectal carcinoma, which is difficult to treat efficiently. The reason why the CD133 (prominin-1) antigen, a surface transmembrane glycoprotein, was chosen for this study is that it is one of the most frequently used cell surface antigens for detection and isolation of cancer stem cells, from various solid tumors [69], including the brain, colon, pancreas, prostate, lung, and liver. Among others, its expression has been documented also for CaCo-2 cells [52]. The prepared PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab in this study demonstrated the proper interaction of the nanoparticles with the CD133 cell surface antigen, which can confirm the delivery of the multicomponent system and can be used as a drug delivery of OXA, as the metabolic assay showed the elimination of the cancer cells.

Synthesis of Iron Oxide Nanoparticles
Synthesis of IO NPs coated with oleic acid (IO-OA NPs) was achieved by co-precipitation using the protocol described in our previous work [33]. Briefly, iron (II) chloride tetrahydrate and iron (III) chloride hexahydrate at a molar ratio of 1:2 were incorporated in degassed Millipore water. To initiate the co-precipitation reaction, ammonium hydroxide was added to the mixture. Oleic acid was added to the reaction system, which was then heated to 80 • C and left for magnetic stirring for 1 h. Thereafter, the IO-OA suspension was washed several times before drying in the oven and redispersed in chloroform. The final concentration of IO-OA in the suspension was set to 10 mg·mL −1 .

Synthesis of PLGA-PEG-NH 2 Copolymer
Synthesis of the amine-terminated diblock copolymer was prepared based on the protocol described in our previous work [33]. In short, PLGA was first activated using NHS and DCC using carbodiimide chemistry. The reaction was left for magnetic stirring for 20 h at room temperature (25 • C). Insoluble dicyclohexyl urea was filtered, followed by precipitation of the polymer using ice-cold diethyl ether. The activated PLGA was dried under vacuum at room temperature (25 • C), followed by the conjugation with the excess of NH 2 -PEG-NH 2 to avoid the formation of triblock copolymer. The amine-terminated diblock copolymer was precipitated and washed with ice-cold methanol. The final PLGA-PEG-NH 2 polymer was dried under vacuum at room temperature (25 • C).

Preparation of the Multicomponent Drug Delivery Systems
The multicomponent drug delivery systems were prepared by the double emulsion method described by Moreno et al. [42], with slight modifications, namely, PLGA_IO-OA_OXA and PEGylated PLGA_IO-OA_OXA. Fifty milligrams of PLGA or PLGA-PEG-NH 2 were dissolved in 1.5 mL of chloroform, followed by the addition of 5 mg of the IO-OA suspension (10 mg·mL −1 ). Oxaliplatin (5 mg) was dissolved in 1 mL of Millipore water and it was added to the polymer/IO-OA organic solution and was emulsified for 5 min at 40 W on an ice-bath using a sonication probe (Sonopuls HD 450, Bandelin) with ON/OFF states of 10 s each. The resulting emulsion was transferred into an aqueous solution of PVA 3% (w/w) (6 mL) and it was emulsified by sonication at 40 W for 5 min with ON/OFF states of 10 s each. The final emulsion was diluted to a final concentration of 0.6 % (w/v) PVA. Finally, the suspension was left for magnetic stirring for 24 h to remove the organic solvent. It was then purified by dialysis for 48 h (molecular weight cut off 12-14 kDa) against 600 mL of Millipore water (with frequent changes of Millipore water). The final nanoparticle suspension was filtered through a 5 µm filter to remove the dust. The final concentration of all nanoparticles was set to 1.8 mg·mL −1 . The ratio of oxaliplatin to polymer and IO-OA to polymer was set to 1:10 (w/w) [33]. It is worth noting that the ratio of IO-OA polymer was set to 1:10 (w/w) based on our previous work, in which the effect of the ratio of IO-OA to polymer on the encapsulation efficiency, mean diameter, and the magnetic properties of the loaded nanoparticles was investigated and optimized [33]. For the sake of fundamental understanding of the behavior of multicomponent loaded delivery systems, polymeric nanoparticles loaded with a single therapeutic agent were also prepared following the same procedure, namely, PLGA_IO-OA, PLGA_OXA, PEGylated PLGA_IO-OA, and PEGylated PLGA_OXA.

Drug Loading and Dissolution Test
The drug loading content which represents the weight fraction of oxaliplatin captured in the nanoparticles after the processing was determined by means of thermogravimetric analysis (TGA, Stanton-Redcroft TG 750). TGA were performed under nitrogen atmosphere (20 mL·min −1 ) from room temperature (25 • C) to 800 • C at a rate of 10 • C·min −1 The encapsulation efficiency (EE (%)) was calculated as follows: To carry out the dissolution tests, 10 mg of the freeze-dried nanoparticles were suspended in 5 mL of PBS (0.1 M, pH 7.4) in a dialysis bag (molecular weight cutoff 3.5 kDa). The suspensions were then incubated in 35 mL of PBS at 37 • C under magnetic stirring at 100 rpm. At specific intervals, 1.5 mL of sample was taken from the PBS bath and refilled with the same volume of fresh PBS. The concentration of released oxaliplatin as a function of time was monitored using UV-visible spectrophotometer at a wavelength of 200 nm based on a calibration curve. All experiments were reproduced in triplicates.

In Vitro Studies in Human Cells
The suspensions of PLGA_OXA_Ab and PLGA_IO-OA_OXA_Ab were sterilized by filtration using a 0.45-µm filter and UV-light irradiation for 20 min. The following cell lines were used for the in vitro experiments: CaCo-2 (human cells derived from colon carcinoma), and MRC-5 cells (human primary noncancerous cells); all cells were from the American Tissue Culture Collection. The cells were cultured in the following cultivation media: CaCo-2 cells in EMEM supplemented with 20% fetal bovine serum (FBS), and MRC-5 cells in EMEM with 10 % FBS. The cells were maintained at the exponential growth phase and, for the experiments, they were seeded on 35-mm microscopic glass-bottom dishes in a sterile hood with laminar flow. Next, 100,000 cells were seeded per one dish in 2 mL of cultivation media and incubated at standard experimental conditions (37 • C, 5 % CO 2 , and 95% humidity) in a sterile incubator. At 16 h after seeding, the cell culture medium was removed and the cells were washed with PBS, which was then exchanged with 2 mL of fresh media with the addition of the tested nanoparticles (PLGA_IO-OA_OXA_Ab) at the following amounts: 2 µL (3.6 µg), 20 µL (36 µg), and 100 µL (180 µg). The stock concentration of the nanoparticles was 1.8 mg·mL −1 . Then, the cells were incubated in the incubator for 30 min and subjected to fluorescence microscopy. In some cases, the cells were stained or co-stained with anti-CD133 conjugated to Atto 565 (2 µL of the antibody per 2 mL of the cell sample for 30 min.). Untreated cells incubated only in media free of nanoparticles served as a negative control.

Sample Fixation
In addition to living cells, fixed samples were also prepared. The procedure corresponds to that described in Section 4.7. Then, after the samples' incubation with the nanoparticles, the medium was removed, the samples were twice washed with PBS, fixed with a 4% solution of methanol-free formaldehyde (Thermo Fisher Scientific, Waltham, MA, USA) in PBS for 20 min at 22 • C in the dark. Then, the samples were washed twice by PBS and kept in dark before fluorescence microscopy.

Fluorescence Microscopy
PLGA_IO-OA_OXA_Ab binding to cells of the examined cell lines was performed by wide-field fluorescence microscopy of both live and fixed samples using an Olympus IX-81 microscope (Olympus, Tokyo, Japan). The images were acquired using EM-CCD camera C9100-02 (Hamamatsu, Japan) by xCellence software. The 60x oil immersion objective with an NA of 1.4 was used. For sample excitation, a high-stability 150 W xenon burner was applied in combination with an excitation/emission FITC and TRITC filter to visualize Alexa Fluor 488 and Atto 565, respectively. The images were background-corrected and deconvolved by two-dimensional deconvolution with the no-neighbor algorithm (Olympus, Tokyo, Japan).

Metabolic Activity Assay
The impact of the prepared nanoparticles with OXA on cell viability was examined in vitro using WST-1 assay (Sigma-Aldrich, St. Louis, MO, USA). For each cell line, 5000 to 10,000 cells (depending on the specific cell line) per well of 96 wells was seeded into single wells in 100 µL of complete cell cultivation media. Then, the cells were cultivated for 16 h at standard cultivation conditions, after which they were treated with 100 µL of fresh cell cultivation media containing the evaluated nanoparticles. Cell viability was then determined 72 h after cell treatment by the following procedure: the culture medium was discarded and changed for 100 µL of fresh phenol red-free DMEM with the addition of 5 µL of WST-1 reagent. After that, the cells were incubated for an additional 2 h in the incubator. Then, the absorbance (of formazan) was measured spectrophotometrically at 450 nm (ref. wavelength of 630 nm) using a UV-Vis spectrometer (Bio-Rad Laboratories, Hercules, CA, USA). All samples were measured in quadruplicates. Untreated cells and cells treated only with blank (pure nanoparticles) served as controls.

Characterization Dynamic Light Scattering
For dynamic light scattering measurement, the particle size distribution was measured using a Zetasizer Nano ZS (Malvern Instruments, Malvern, Great Britain). The amount of 1 mL of each suspension was placed in a cuvette without further dilution. All measurements were performed in triplicate in an autocorrelation mode with a single illuminating beam and detector with a scattering angle of 90 • at room temperature (25 • C).

Zetasizer Potential
The zeta potential of the prepared nanoparticles was measured using Zetasizer Nano ZS (Malvern Instruments, Malvern, Great Britain) using the Smoluchowski model. The measurements were carried out on 800 µL of 10× times diluted aqueous mother suspension. Measurement of the zeta potential of IO-OA was performed by dispersing 28 µL of IO-OA suspension in 1 mL of Milli-Q water and the removal of chloroform by evaporation before the analysis. It is worth noting that the colloidal stability of IO-OA in an aqueous solution was retained during the measurement. All measurements were performed in triplicates at room temperature (25 • C).

Transmission Electron Microscopy
The morphology of the prepared nanoparticles was characterized by Transmission Electron Microscopy (TEM). The measurements were carried out using 100 kV TEM, model JEM-1010 (JEOL, Ltd.) equipped with CCD camera MegaView III (Olympus Soft Imaging Systems, Germany). In short, 10 µL of suspension was deposited on a carbon-coated electron microscopic grid. Sample contrasting was performed using 1% (w/w) uranyl acetate. The grid was dried and then was introduced into the electron microscope column for analysis.

Attenuated Total Reflection Fourier-Transformed Infrared Spectroscopy
Thermo Scientific Nicolet iZ10 Fourier transform infrared spectrometer in attenuated total reflection mode (ATR-FTIR) with a diamond crystal was utilized to identify the molecular structure of the PLGA based copolymers and the drug delivery systems. The FTIR spectra are an average of 64 scans.

Thermogravimetric Analysis
Thermogravimetric analyses (TGA) were carried out using Stanton Redcroft TG 750 under nitrogen atmosphere (20 mL·min −1 ) from room temperature (25 • C) to 800 • C at a rate of 10 • C·min −1 to determine the drug loading content of OXA captured in the nanoparticles during the processing.

Vibrating Sample Magnetometer
The magnetic properties of the multicomponent drug delivery systems were assessed by a vibrating sample magnetometer (VSM 7407, Lake Shore, MD, USA) at room temperature (25 • C) in the magnetic field from −10 kOe to +10 kOe.

Statistical Analysis
All measurements were performed in triplicates and were expressed as means with standard deviation. Differences between nanoparticles were tested using a one-way ANOVA with a level of significance of p ≤ 0.05.

Conclusions
In this study, Ab conjugated PLGA nanocarriers for the targeted delivery of OXA and superparamagnetic nanoparticles to colorectal cancer cells were developed with the aim of suppressing the IPN caused by cumulative doses of OXA, and to enhance the treatment efficacy. Co-encapsulation of OXA and IO-OA was achieved in two polymeric carriers, namely, PLGA and PLGA-PEG, using the double emulsion method. PLGA_IO-OA_OXA and PE-Gylated PLGA_IO-OA_OXA displayed a comparable mean diameter of 207 ± 70 nm and 185 ± 119 nm, respectively. The encapsulation of OXA and/or IO-OA and the PEGylation process did not statistically affect the hydrodynamic size of the nanoparticles. The concentration of the released OXA from the PEGylated PLGA_IO-OA_OXA increased very rapidly, reaching~100% release after only 2 h, while the PLGA_IO-OA_OXA displayed a slower and sustained drug release. For a controlled release of OXA, non-PEGylated PLGA nanoparticles are thus more suitable. Preservation of the superparamagnetic behavior of the IO-OA without magnetic hysteresis all along the dissolution process was also pointed out, thereby making the multicomponent drug delivery systems suitable for treatment with magnetically mediated hyperthermia, even after the drug release is complete.
To specifically target cancer cells overexpressing CD133, the non-PEGylated nanoparticles were conjugated to anti-CD133 Alexa Fluor 488 Ab. The Ab-coated nanoparticles were efficiently bound to CD133 expressing cells, as documented in the CaCo-2 cells. The OXA-containing nanoparticles were able to eliminate the cancer cells, which was shown using a metabolic activity assay. Therefore, the prepared nanoparticles are suitable not only for specific cancer cell targeting and selective drug delivery, but also for the imaging of such cells and their eradication. The concept is expandable and could also be used for the delivery of other anticancer drugs.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Acknowledgments:
The authors are also grateful to František Štepánek (UCT Prague) for the access to the Zeta-sizer and FTIR equipment.