Effects of Electromagnets on Bovine Corneal Endothelial Cells Treated with Dendrimer Functionalized Magnetic Nanoparticles

To improve bovine corneal endothelial cell (BCEC) migration, enhance cell energy, and facilitate symmetric cell distribution in corneal surfaces, an electromagnet device was fabricated. Twenty nanometer superparamagnetic iron oxide nanoparticles (SPIONs) functionalized with fourth-generation dendrimer macromolecules were synthesized, and their size and structure were evaluated using transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and Fourier transform infrared spectroscopy (FTIR). The results confirmed the configuration of the dendrimer on the SPION surfaces. In vitro biocompatibility was assessed using the 3-[4,5-dimethylthiazol-2yl]-2,5-diphenyl tetrazolium bromide assay. No significant toxicity was noted on BCECs within 24 h of incubation. In the cell migration assay, cells treated with dendrimer-coated SPIONs exhibited a relatively high wound healing rate under sample addition (1 μg/mL) under a magnetic field. Real-time PCR on BCECs treated with dendrimer-coated SPIONs revealed upregulation of specific genes, including AT1P1 and NCAM1, for BCECs-dendrimer-coated SPIONs under a magnetic field. The three-dimensional dispersion of BCECs containing dendrimer-coated SPIONs under a magnetic field was evaluated using COMSOL Multiphysics software. The results revealed the BCECs-SPION vortex pattern layers in the corneal surface corresponded to the electromagnet’s displacement from the ocular surface. Magnetic resonance imaging (MRI) indicated that dendrimer-coated SPIONs can be used as a T2 contrast agent.


Synthesis of SPIONs
The SPIONs were synthesized by thermal decomposition of the Fe(acac) 3 inside a solution containing oleic acid, oleylamine, and octyl ether. The mixture was stirred for 30 min at room temperature under a nitrogen flow. Next, the temperature was increased to 300 • C for 1 h until the Fe(acac) 3 had been completely decomposed. Finally, the solution was cooled to room temperature, washed several times with ethanol, and collected using permanent magnets [21].

D-SPIONs
The PAMAM-functionalized SPIONs were fabricated through three reaction steps, as described previously [22]. During the fabrication process, SPION surfaces should first be modified by amine group attachment for subsequent dendrimer coating. In brief, APTES in ethanol (10 mL, 10% vol/vol) was added dropwise to the SPIONs dispersed in ethanol through sonication. The mixture was stirred for 7 h, separated by magnets, washed with ethanol five times, and collected for further analysis. For the synthesis of the PAMAMcoated SPIONs, the methyl acrylate (15 mL, 20% vol/vol) was dissolved in methanol and added to the SPIONs functionalized with aminosilane (20 mL, 5% wt/vol) and stirred for 48 h. This was followed by 2 h of sonication. Next, the SPIONs were washed five times. Finally, ethylenediamine was dissolved in methanol (10 mL, 50% vol/vol), and added to the modified SPIONs, and the mixture was sonicated for 3 h. The resulting firstgeneration dendrimer on the SPION surface was then subjected to washing and purification using magnets. The two main reactions, involving the generation of ester bonds through the connection of methyl acrylate and amino silane in the first step and followed by the amidation of ester bonds with ethylenediamine in the second step, was repeated three times to fabricate the fourth-generation dendrimer branches ( Figure 1). purification using magnets. The two main reactions, involving the generation of ester bonds through the connection of methyl acrylate and amino silane in the first step and followed by the amidation of ester bonds with ethylenediamine in the second step, was repeated three times to fabricate the fourth-generation dendrimer branches ( Figure 1).  A field-emission Tecnai F20 electron gun of ZrO/W (100) Schottky type (resolution ≤0.23 nm) was used on a (Philips/FEI Corporation, Eindhoven, The Netherlands) TEM system. XPS was performed using an Al Kα X-ray source under a working energy of 1486.6 eV in a vacuum of approximately 10 −7 Pa (XPS, ESCA000600). FTIR spectroscopy was conducted by grinding the SPIONs and D-SPIONs into powder with the KBr to make pellets using a Perkin Elmer system (PerkinElmer, Inc. Waltham, MA 02451 USA) in the range of 400-4000 cm −1 . The DLS and zeta potential of the SPIONs and D-SPIONs dispersed in deionized water were measured using Malvern instrument (Malvern Panalytical Ltd, Great Malvern, UK).

Bovine Corneal Endothelial Cell Preparation
The bovine eyes were obtained from a local butchery and fumigated using an iodine solution [23]. After the eyes were washed with phosphate-buffered saline (PBS), the corneas were removed, as were the Descemet membranes. Subsequently, the Descemet membranes were incubated with trypsin under a CO2 atmosphere at 37 °C for 45 min until the cells were detached. After centrifugation at 1000 rpm for 5 min, the BCECs were collected and cultured.  A field-emission Tecnai F20 electron gun of ZrO/W (100) Schottky type (resolution ≤0.23 nm) was used on a (Philips/FEI Corporation, Eindhoven, The Netherlands) TEM system. XPS was performed using an Al Kα X-ray source under a working energy of 1486.6 eV in a vacuum of approximately 10 −7 Pa (XPS, ESCA000600). FTIR spectroscopy was conducted by grinding the SPIONs and D-SPIONs into powder with the KBr to make pellets using a Perkin Elmer system (PerkinElmer, Inc. Waltham, MA 02451 USA) in the range of 400-4000 cm −1 . The DLS and zeta potential of the SPIONs and D-SPIONs dispersed in deionized water were measured using Malvern instrument (Malvern Panalytical Ltd, Great Malvern, UK).

Bovine Corneal Endothelial Cell Preparation
The bovine eyes were obtained from a local butchery and fumigated using an iodine solution [23]. After the eyes were washed with phosphate-buffered saline (PBS), the corneas were removed, as were the Descemet membranes. Subsequently, the Descemet membranes were incubated with trypsin under a CO 2 atmosphere at 37 • C for 45 min until the cells were detached. After centrifugation at 1000 rpm for 5 min, the BCECs were collected and cultured.

MTT Cell Viability Assay
The 3-[4,5-dimethylthiazol-2yl]-2,5-diphenyl tetrazolium bromide (MTT) (Sigma Aldrich, St. Louis, MO, USA). viability assay of D-SPIONs was performed using bovine CECs (BCECs) for the duration of 24, 48, and 72 h. For the MTT assay, the 6 × 10 3 cells/well are cultured and grown on 96-well plates at 37 • C in an endothelial cell medium and 1% penicillin for 24 h. The cells were then incubated with five concentrations of D-SPIONs (10,30,50,70, and 100 µg/mL) for 24, 48, and 72 h. At the end of the incubation periods, 10 µL of MTT was added to each well, and the cells were kept in the incubator for 3 h. Next, the cells were washed with PBS, 100 µL of dimethyl sulfoxide was added to each well, and the plate was shaken for 10 min. Finally, the optical density of each cell was determined at 595 nm on a microplate reader (Bio-Rad S/N 21648, Pleasanton, CA, USA) Using the absorbance data, the percentage of live cells was calculated [24,25].

MRI of Fourth-Generation D-SPIONs
To evaluate the MRI relaxivity of the fourth-generation D-SPIONs as T 2 -weighted contrast agents, a 7 Tesla Bruker BioSpec MRI system (Bruker BioS pec 70/30 US, Billerica, MA, USA) was used. T 2 -weighted relaxation times were defined by a multiecho spin-echo sequence (repetition time: 4000 ms; echo time:18 ms). The r 2 relaxivity was collected from the slope of 1/T 2 versus the D-SPION concentrations, and signal decay data were collected. Phantoms were prepared in varying concentrations (0.1, 0.2, 0.3, 0.5, 0.7, and 1 mg/mL) by adding 1% agarose. The samples were then transferred to 0.5 mL microtubes and placed inside a 7 cm coil. Images were taken using the designed sequences with a matrix size of 128 × 128, a field of view of 6 × 6 cm 2 , and a slice thickness of 1 mm.

Electromagnet Setup
The electromagnet device consisted of two electromagnets facing each other under an up/down adjustable magnetic field gradient. The magnets were placed under a step functional magnetic field of approximately 500 G every 2 s and were triggered by using the Arduino device, delivering a step-functional ± 5 volts. The magnetic field consisted of two commercial electromagnets (Grove, seeed studio, Shenzhen, China) which were made by a thousand-rounds wire and a cold-rolled steel rod, and the working gap was tunable. The strength of the magnetic field was measured using a Tesla meter (TM-801 EXP, KANETEC) and was sensed at the bottom position of the dish [26].
After the BCECs were cultured and the number of cells reached the experimental requirements (10 4 BCECs in each 96-well plate), D-SPIONs (10 µg/mL) were added to the dish containing the cells and incubated for 24 h such that they could be absorbed by the cells. Finally, the BCECs were washed with PBS, and the D-SPION cells were placed under an applied magnetic field in an incubator for 1 month (Figure 6a).

BCEC Migration Assay
An in vitro BCEC scratch assay was performed. In brief, BCECs were seeded in a 96-well plate at a density of 10 4 cells/well and incubated for 24 h. Next, two concentrations of dendrimer coated iron oxide (1 and 10 µg/mL) were added to each well and incubated for 24 h at 37 • C under a CO 2 atmosphere (untreated BCECs were used as controls). Next, straight lines are scratched in each well using a sterile 100 µL pipette tip. Images of the scratches in each well were examined at 100 × magnification under an inverted microscope and analyzed with ImageJ software. Each experiment was performed twice. The scratch experiments were repeated for the cell migration assay for the same concentrations under a magnetic field produced by electromagnets in tapping modes.

Simulation Study
A 3D simulation study was performed by COMSOL (COMSOL, Inc. Burlington, MA, USA), to confirm the distribution of the BCECs holding nanoparticles on the ocular surface. The uniform distribution of the CECs on the ocular surface is challenging for gene delivery and drug delivery toward the cornea. The delivery to CECs containing SPIONs was determined to be nonuniform, creating cell accumulations on the ocular surface. Herein, the density of the D-SPIONs was less than 10 13 (1/m 3 ); therefore, their contribution was ignored in the simulation. According to Maxwell's equation, the magnetic flux density of the magnet can be expressed as: where the µ 0 denotes permeability, H is the magnetic field intensity, and M represents the magnetization. As presented in Figure 7, the cylindrical permanent magnet radius is labeled as r m , the distance between the magnet and the eyeball was denoted as D, and the eyeball radius was denoted as r. Physical properties of each item used in simulation are listed in Table 1. Real-time PCR was performed according to the manufacturer's protocol. The cDNA was synthesized using RNA, a random hexamer primer, Revert Aid Reverse Transcriptase, and a dNTP mix. The cDNA of the treated and untreated D-SPION cells were diluted using deionized water.
Real-time PCR was conducted using the Applied Biosystems 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The 20 µL container was filled with cDNA and the as prepared solution contained SYBR Green Real-Time PCR Master Mix, gene-specific primers, and deionized water. The gene-specific primers are listed in Table 2. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used to normalize the gene expression values, and melt curve analysis was conducted to assess the reaction specifications [27,28]. The 2 −∆∆Ct method was employed to analyze the data. The cycle thresholds for the intrinsic control of GAPDH were derived from the cycle thresholds for all the cell markers. The fold changes for the D-SPION cells under a magnetic field were normalized to those of the untreated cells, and the values underwent an exponential linear conversion by using the 2 −∆∆Ct equation [29]. Table 2. Primers used in this study.

Gene
Gene Sequence Analyses were conducted using SPSS software, and one-way analysis of variance was performed to find differences between SPIONs and D-SPIONs. A p-value of ≤0.05 was considered significant. Data are presented as ± standard errors of the mean.

Shape, Size, Surface Charge, and Structural Bonding of D-SPIONs
The shape, size, Surface charge, and structural bonding of D-SPIONs are shown in Figure 2.
confirmed that the dendrimer molecules were attached to the iron oxide surfaces [31]. The SPION peak Fe2p and the oxygen peak O1s exhibit a reduction in their intensity by covering the SPIONs with the PAMAM layers, but the amount of the carbon (C1s) atoms is increased, and the nitrogen peaks appear in the D-SPION spectra. The increase in dendrimer layers at the SPION surface reduces the amount of the iron and oxygen molecules through a process wherein the dendrimer thickness is increased [32].

FTIR
The attachment of the fourth-generation dendrimers on SPION surfaces was confirmed using Fourier transform infrared (FTIR) spectroscopy. As presented in Figure 2e, the sharp peak at 590 cm −1 was assigned to the Fe-O stretching vibration of the iron oxide nanoparticles at their tetrahedral site bond. As presented in the D-SPION spectra, the peaks at 1489 and 1567 cm −1 were attributed to the -CO-NH-groups, and the peak at 1023 cm −1 was ascribed to the -CO-NH 2 groups, which are highly electronegative. Both peaks confirm the existence of the dendrimer branch on the SPION surfaces. Moreover, the peak at 3409 cm −1 was related to the -NH 2 -group vibration of the dendrimer [30]. Overall, the FTIR spectra indicate that the dendrimer modified the SPIONs.

XPS
Surface analysis of the SPIONs and D-SPIONs (G 4 ) was performed using XPS (Figure 2f). The peak in the SPION spectra corresponded to the C1s (285 eV), O1s (528.5 eV), and Fe2p (710.3 eV). In the D-SPION spectra, a peak corresponding to the nitrogen sign at N1s (398.2 eV) was observed [2]. XPS analysis revealed the presence of the dendrimer on the iron oxide nanoparticle surfaces. The peak position of N1s, which appeared at 390-400 eV, confirmed that the dendrimer molecules were attached to the iron oxide surfaces [31]. The SPION peak Fe2p and the oxygen peak O1s exhibit a reduction in their intensity by covering the SPIONs with the PAMAM layers, but the amount of the carbon (C1s) atoms is increased, and the nitrogen peaks appear in the D-SPION spectra. The increase in dendrimer layers at the SPION surface reduces the amount of the iron and oxygen molecules through a process wherein the dendrimer thickness is increased [32].

DLS and zeta potential
The hydrodynamic size of SPIONs and the fourth-generation D-SPIONs were measured using DLS. As shown in Figure 2g, the hydrodynamic size of the SPIONs was increased after they were coated with the dendrimer. The hydrodynamic diameter of the SPIONs and D-SPIONs exceed the actual sizes seen in the TEM images; this inaccuracy is the main disadvantage of using DLS. In DLS, signals from larger particles can suppress signals from smaller particles, and it could be associated with the higher contribution of the light scattering to the nanoparticle size distributions. This means that higher size distributions can lead to errors in DLS size measurements, yielding discrepancies with the real size indicated in TEM. Furthermore, for highly magnetic nanoparticles, nanoparticle agglomeration occurs during measurement, which can affect the analytical signals. The surface charge study of SPIONs and D-SPIONs, as performed through zeta potential measurement, was −24 and −0.6 mV, respectively. These results can be attributed to the carboxylate moiety on the SPION surface. This surface is further functionalized by the amine groups of PAMAM producing D-SPIONs [33].

MTT Cytotoxicity Assay
The 3-[4,5-dimethylthiazol-2yl]-2,5-diphenyl tetrazolium bromide (MTT) cytotoxicity assay to investigate the cytotoxicity of five concentrations of D-SPIONs under 24, 48, and 72 h of treatment on BCECs (Figure 3). The BCECs exhibited various biological responses when exposed to differing dosages of D-SPIONs over varying durations. No significant toxicity of D-SPIONs on BCECs was noted up to 100 µg/mL for 24 h. The cell viability for the 100 µg/mL under 48 and 72 h of incubation was 68% and 65%, respectively. Because iron oxide nanoparticles are biocompatible and nontoxic, they display slight cytotoxicity against BCECs [28,34]. and 72 h of treatment on BCECs (Figure 3). The BCECs exhibited various biological responses when exposed to differing dosages of D-SPIONs over varying durations. No significant toxicity of D-SPIONs on BCECs was noted up to 100 μgmL for 24 h. The cell viability for the 100 μgmL under 48 and 72 h of incubation was 68% and 65%, respectively. Because iron oxide nanoparticles are biocompatible and nontoxic, they display slight cytotoxicity against BCECs [28,34].   Figure 4 presents the MRI images of the D-SPIONs specifically, T 2 images at various sample concentrations (Figure 4a) and relaxation rates against various sample concentrations (Figure 4b). Dose dependence was noted, as indicated by the darkening that occurred as the spin-spin relaxation time (T 2 ) was shortened. The transverse relaxation rate (1/T 2 ) versus the various concentrations of the D-SPIONs is presented in a linear plot [35]. The relaxivity r 2 was calculated to be 46 mM −1 S −1 .The SPIONs and D-SPIONs were reported to enhance the transverse relaxation rate more than the longitudinal relaxation rate and be a negative contrast agent. Dendrimer-based MRI contrast agents has also been used for imaging brain tumors [36]. As synthetic macromolecules, dendrimers can improve the relaxivity in MRI because of their 3D structures, tunable sizes, and abundant surface terminals. Furthermore, they can increase the positivity of iron oxide. Coating dendrimers on the SPION surface promote their specific targeting, enhance MRI image quality, reduce their toxicity on cells, and allow for safe in vivo imaging.

Cell Migration Assay under Magnetic Fields
Migration in CECs is an essential step in the wound healing process. Cell migration was suppressed in the BCECs treated with D-SPIONs compared with in the control cells (BCECs without nanoparticles). Moreover, the migration of the BCECs containing 1 µg/mL D-SPIONs was faster than that of the cells containing 10 µg/mL SPIONs ( Figure 5). Increases in nanoparticle agglomeration, cell uptake capacities, and nanoparticle moieties on the cell surfaces are the main reasons for the delay in cell migration under increases in the D-SPION concentration [37]. Under a magnetic field produced by electromagnets in tapping mode, wound closure in the BCECs treated with 1 µg/mL SPIONs took 10 h. This is even faster than wound closure in the control BCECs. Notably, the magnetic vortex patterns were created by BCECs containing D-SPIONs by applying the magnetic field for both concentrations of D-SPIONs (1 and 10 µg/mL) at different times (Figure 6b) [38]. contrast agents has also been used for imaging brain tumors [36]. As synthetic macromolecules, dendrimers can improve the relaxivity in MRI because of their 3D structures, tunable sizes, and abundant surface terminals. Furthermore, they can increase the positivity of iron oxide. Coating dendrimers on the SPION surface promote their specific targeting, enhance MRI image quality, reduce their toxicity on cells, and allow for safe in vivo imaging.

Cell Migration Assay under Magnetic Fields
Migration in CECs is an essential step in the wound healing process. Cell migration was suppressed in the BCECs treated with D-SPIONs compared with in the control cells (BCECs without nanoparticles). Moreover, the migration of the BCECs containing 1μg/mL D-SPIONs was faster than that of the cells containing 10 μg/mL SPIONs ( Figure 5). Increases in nanoparticle agglomeration, cell uptake capacities, and nanoparticle moieties on the cell surfaces are the main reasons for the delay in cell migration under increases in the D-SPION concentration [37]. Under a magnetic field produced by electromagnets in tapping mode, wound closure in the BCECs treated with 1 μg/mL SPIONs took 10 h. This is even faster than wound closure in the control BCECs. Notably, the magnetic vortex

COMSOL Multiphysics Simulation Study
In the simulation, we assumed no current conveyance in the domain (H = 0); next, the magnetic flux density was adjusted through the magnet. The magnetization was calculated to be 10 4 (A/m) in the z direction from the electromagnet under a r m of 0.8r and a D of 0.2r. Table 1 lists the other parameters. The magnetic flux density created on the ocular surface was inversely proportional to the magnet's displacement from the ocular surface ( Figure 7). When the displacement was increased, the magnetic flux density decreased, affecting the number of magnetic patterns on the ocular surface. In this regard, the isometric line vortex around the magnetization direction was considered.
Polymers 2021, 13, x FOR PEER REVIEW 11 of 15 In the simulation, we assumed no current conveyance in the domain (H = 0); next, the magnetic flux density was adjusted through the magnet. The magnetization was calculated to be 10 4 (A/m) in the z direction from the electromagnet under a rm of 0.8r and a D of 0.2r. Table 1 lists the other parameters. The magnetic flux density created on the ocular surface was inversely proportional to the magnet's displacement from the ocular surface ( Figure 7). When the displacement was increased, the magnetic flux density decreased, affecting the number of magnetic patterns on the ocular surface. In this regard, the isometric line vortex around the magnetization direction was considered.

Real-Time PCR
To investigate the change in the cellular identity of the BCECs before and after the treatments with the alternative magnetic field, real-time PCR was conducted (Figure 8).

Real-Time PCR
To investigate the change in the cellular identity of the BCECs before and after the treatments with the alternative magnetic field, real-time PCR was conducted ( Figure 8). To measure the mRNA expression of specific genes, BCEC-specific markers (ATP1A1, CDH2, ENO2, NCAM1, SLC4A4, and zonula occludens-1 [ZO-1]), were used for untreated cells and treated cells for 1 month [39,40]. The markers were selected based on strong expression of genes in BCECs, and low expression of them in stromal cells. Comparison of the gene expression of D-SPION endothelial cells before and after 1-month cell treatment revealed an approximate sixfold growth in ATP1A1 and NCAM1 expression and an approximate 2.5-fold growth in ENO2 and SLC4A4. No major changes were observed in ZO-1 and CDH 2 expression; moreover, no clear differences in the tight junction morphology of the BCECs before and after the treatments (ZO-1 is a protein involved in cellular function). The significantly high increase in adenine triphosphatase (ATPase)Na+/K+ transporting α1polypeptide (ATP1A1) was reported. ATP1A1 supplies instructions for fabricating the alpha-1 subunit of a Na + /K + ATPase, which is known as a protein pump. ATP is then used by the protein to transport K + ions into and Na + ions out of cells [41]. The solute carrier 4 SLC4 transporters (except SLC4A11) are divided into three groups based on their functions: anion exchangers, sodium bicarbonate cotransporters (NBCs), and sodiumdriven Cl-/HCO-exchangers [42,43]. Next, the expression of SLC4A4 as an electrogenic NBC1 was localized in the basolateral membrane of corneal epithelial cells. The expression of the SLC4A4 gene (as a primary member in the corneal endothelium) suggests that NBCs are the main bicarbonate transporters in the corneal endothelium. Moreover, for the SLC4A4 gene upregulated, the increase of salt content in the medium after magnetic treatments could increase the expression of solute transporter [44]. The expression of the neural cell adhesion molecule, which is encoded by the NCM1 gene, was significantly increased. NCM, an adhesion molecule known as the immunoglobulin gene [45], can interface neuron-muscular and neuron-neuron adhesion through their interactions and initiate cell matrix interactions. NCAM expression by HCECs has a neural crest origin. NCAM overexpression might correspond to the adhesion of the BCECs to each other.
Polymers 2021, 13, x FOR PEER REVIEW 12 of 15 was significantly increased. NCM, an adhesion molecule known as the immunoglobulin gene [45], can interface neuron-muscular and neuron-neuron adhesion through their interactions and initiate cell matrix interactions. NCAM expression by HCECs has a neural crest origin. NCAM overexpression might correspond to the adhesion of the BCECs to each other.

Conclusion
The SPION surfaces were functionalized with the fourth-generation dendrimer for better spatial distance. The TEM images revealed that the D-SPIONs measured approximately 20 nm. FTIR spectroscopy and XPS confirmed the coating of the dendrimer on the SPION surfaces. The zeta potential of the SPIONs changed from approximately −28 to −0.6 mV after dendrimer functionalization because of the induction of amine groups on its surfaces. The MTT assay indicated that the D-SPION had low toxicity on the BCECs over 24 h. Moreover, the coated nanoparticles could be used as contrast agents for in vivo bi- Figure 8. Gene expression level of BCECs incubated with D-SPIONs and treated with an electromagnet system using real time PCR. The genes were normalized using GAPDH. The cells incubated with the MNPs were used as the control group and statistically differences were (p < 0.05; n = 3). Data are presented as the means standard deviations of three experiments.

Conclusions
The SPION surfaces were functionalized with the fourth-generation dendrimer for better spatial distance. The TEM images revealed that the D-SPIONs measured approxi-