Induction of Heat Shock Protein-72 by Magnetic Nanofluid Hyperthermia in Cultured Retinal Ganglion Cells for Neuroprotective Treatment in Glaucoma

1Department of Ophthalmology, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea 2Biomagnetics Laboratory (BML), Department of Electrical and Computer Engineering, National University of Singapore, Singapore 117576 3Department of Neurosurgery, Ischemic/Hypoxic Disease Institute, Biomedical Research Institute, and Cancer Research Institute, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 110-744, Republic of Korea


Introduction
Glaucoma is characterized by progressive optic nerve damage with selective loss of retinal ganglion cells (RGCs) [1][2][3].Recent clinical trials confirmed that the major modifiable risk factor for the development and progression of glaucoma is the intraocular pressure (IOP).The current treatment of glaucoma mainly involves reducing IOP and diurnal fluctuation.However, signs of glaucomatous progression can be seen in many patients with well-controlled IOP.In addition, disease onset and progression can occur in patients whose untreated IOPs are within the normal range.From this background, the importance of neuroprotection in the treatment of glaucoma has become a worldwide issue.During the past decades, many studies have focused on discovering neuroprotective agents along with IOP lowering medications in order to prevent retinal ganglion cell (RGC) death or even reverse the process of cell death [4].
Heat shock proteins (HSPs), also known as stress proteins, are a group of proteins found in all living cells.It is rapidly induced by a variety of environmental stresses, such as hyperthermia, hypoxia, or ischemia, and its role is to protect cells against stress [5,6].This protection mainly attributes to members of the HSP70 family, especially 72 kDa heat shock protein (HSP72).HSP72, the major inducible form of the HSP70 family, has been found protecting cells from certain apoptotic stimuli such as oxidative stress, hypoxia, and inflammation [7][8][9].
Magnetic hyperthermia using superparamagnetic nanoparticle (SPNP) agents is considered a promising biotechnological approach to induce HSPs in a target tissue because it can generate accurately controllable localized heating [10,11].Recently, our group reported on the feasibility of achieving neuroprotection with magnetically engineered superparamagnetic Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles for localized HSP induction [12].However, no empirical study on HSP expression in neuronal cells induced by magnetically engineered SPNP agents has been presented.The purpose of this study is to demonstrate induction of HSPs in cultured RGCs by local magnetic hyperthermia using engineered superparamagnetic Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticle agents.

2.2.
Synthesis of Mn 0.5 Zn 0.5 Fe 2 O 4 and Coating with PEG 500 2.2.1.Synthesis of Mn 0.5 Zn 0.5 Fe 2 O 4 .Mn 0.5 Zn 0.5 Fe 2 O 4 was synthesized using a high temperature thermal decomposition method.Mn (0.5 mmol), Zn (0.5 mmol), Fe (III) (2 mmol), oleic acid (6 mmol), oleylamine (6 mmol), 1.2 hexadecanediol (10 mmol), and benzyl ether (20 mL) were mixed and gently stirred at room temperature.The ramping rate and heat treatment time were 13.3 ∘ C/min and 45 min, respectively.After heat treatment, the mixed solution was cooled to room temperature.Ethanol (40 mL) was added to the mixed solution to rinse the synthesized SPNPs, which were collected by centrifugation and then dried at room temperature.

2.2.2.
Coating of Mn 0.5 Zn 0.5 Fe 2 O 4 with PEG 500.The synthesized Mn 0.5 Zn 0.5 Fe 2 O 4 SPNAs were coated with PEG 500 Da to form a nanofluid.The SPNPs were dispersed in toluene (solvent, 7.5 mL), and then PEG (0.75 mL) and triethylamine (catalyst, 3.75 mL) were added.The mixed suspension was shaken for 24 hours at room temperature.The PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 SPNPs were then collected by centrifugation and dispersed in distilled water.

In Vitro Cytotoxicity Assay: Analysis by Transmission
Electron Microscopy (TEM).TEM analysis was conducted for the RGC-5 cells treated with Mn 0.5 Zn 0.5 Fe 2 O 4 particles to investigate the degree of SPNP penetration into the cells, cell apoptosis, and cell deformation, including nuclear fragmentation.RGC-5 cells were differentiated with 1 M staurosporine for 1 hour and incubated with 300 g/mL SPNP for 24 hours.Cells were fixed with 2.5% glutaraldehyde and washed with PBS followed by 0.1 M cacodylate buffer (pH 7.2) and postfixed with 2% osmium tetroxide in 0.1 M cacodylate buffer for 1.5 hours at room temperature.The samples were then washed briefly with distilled water, dehydrated in ethanol, infiltrated with propylene oxide (Acros Organics) and EPON mixed with epoxy resin (Embed 812, nadic methyl anhydride, Poly Med@ 812, dodecenyl succinic anhydride, dimethylaminomethyl phenol, Electron Microscopy Polysciences), and finally embedded with epoxy resin alone.The epoxy resin mixed samples were loaded into capsules and polymerized at 38 ∘ C for 12 hours and 60 ∘ C for 48 hours.Thin sections were made using an RMC MT-XL ultramicrotome, collected on a copper grid, and viewed with a JEOL (KEM-1400) at 80 kV in the TEM mode.

Physical and Chemical
Characterization of Mn 0.5 Zn 0.5 Fe 2 O 4 SPNP Hyperthermia Agents.The shape, size, coating thickness, and size distribution of SPNPs were analyzed by highresolution transmission electron microscopy (HR-TEM).The  AC,mag of PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 SPNPs was characterized using an AC solenoid coil-capacitor system that we developed.The AC magnetically induced heating temperature of RGCs treated with the SPNPs (500 g/mL) was measured at a fixed AC magnetic field of  = 140 kHz and  = 140 Oe.Heating stress was applied for times ranging from 600 to 1200 s.In addition, measurements were taken of the heating characteristics of RGCs treated with SPNPs at concentrations from 300 to 700 g/mL under the same AC magnetic heating conditions.

Induction of Heat Shock Protein by the Magnetic Nanofluid
Hyperthermia System.The induction of HSPs by AC magnetic nanofluid hyperthermia was conducted using the high performance PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 SPNPs that we developed.In order to investigate the AC magnetically induced heating characteristics in RGCs, we developed an AC magnetic field generation system, which consists of AC coils, capacitors, DC power supplies, and wave generators (see Figure 1).This system was designed for both in vitro and in vivo pilot studies.The applied frequency of this system can be automatically controlled from 30 to 370 kHz, and the applied AC magnetic field can be simultaneously changed from 0 to 200 Oe at each applied frequency.
Before magnetic hyperthermia, the RGC-5 cells were differentiated with 1 M staurosporine for 1 hour and incubated with Mn 0.5 Zn 0.5 Fe 2 O 4 SPNPs for 24 hours.The cells were then washed with PBS (phosphate buffered saline) and treated with trypsin-EDTA (Life Technologies, USA) followed by addition of an equal volume of culture medium.Cells were centrifuged for 5 min; the supernatant was discarded and resuspended in culture medium.After a second 5-minute centrifugation, the supernatant was again discarded and the cells were harvested into 0.1 mL of the medium in a 0.6 mL Eppendorf (E) tube.
After maintaining the temperature of the cells in the Etube at 37 ∘ C in a temperature-controlled bath, we placed the E-tube and cells in the center of the magnetic coil of an induction generator, monitoring the temperature while the generator operated.At the fixed applied frequency of 140 kHz, we controlled the temperature by increasing or decreasing the strength of the AC magnetic field.We maintained the target temperature of 41.0 ∘ C for the desired period before stopping the generator.
The RGCs incubated with Mn 0.5 Zn 0.5 Fe 2 O 4 SPNPs were centrifuged to form a cell pellet.The tube and pelleted RGCs were immersed in a water bath to keep the temperature in the 36-37 ∘ C range.As shown in Figure 1, the temperature of the RGC-5 cells during hyperthermia was measured using an optical thermosensor (Fluoroptic temperature probe, Lumasense technologies, USA).The microcentrifuge tube with an optical thermometer inserted in the RGCs was placed at the center of the coil supplying the AC magnetic field.The warming rate of the SPNPs in the RGCs was controlled by changing the strength of  appl within the biologically and physiologically safe range of 120 Oe ( appl ⋅  appl = 1.34 × 10 9 A m −1 s −1 )-160 Oe ( appl ⋅  appl = 1.78 × 10 9 A m −1 s −1 ) at the fixed frequency of 140 kHz.The  AC,mag of RGCs with SPNPs during HSPs induction was kept at 40.5 ∘ C ± 0.5 ∘ C for 900 sec.
After 24 hours, the RGCs were counted and seeded onto a culture plate followed by fixation with 10% formaldehyde.The RGCs were washed twice with PBST (phosphate buffered saline tween-20) and then blocked.Anti-HSP72 (Enzo Life Sciences, USA) was added as primary antibody, followed by two washes with PBST.Alexa Fluor 488 goat anti-mouse IgG (FITC, fluorescein isothiocyanate) (Life Technologies, USA) was then added as secondary antibody, followed by two washes with PBST.To stain the nuclei, DAPI (4  ,6-diamino-2-phenylindole) (Sigma, USA) was added, followed by two washes with PBST.

Western Blot Analysis.
To quantify the induction of HSPs by hyperthermia using SPNPs, we performed Western blot analyses for four samples: three types of control group and an experimental sample with SPNPs (500 g/mL) and a magnetic field ( appl = 140 kHz and  appl = 140 Oe) applied for 900 sec.For the analysis, harvested cells were lysed in a lysis buffer (Intron Biotechnology, Korea), loaded onto a 10% SDS-polyacrylamide gel, and transferred to a polyvinylidene fluoride membrane.The membrane was blocked with dried nonfat milk and incubated with the primary antibody: Hsp72 antibody (1 : 1000, Enzo Life Sciences, USA) and -actin antibody (1 : 1000, Sigma, USA).The membrane was washed with Tris-buffered saline containing 0.1% tween-20 and incubated with goat anti-mouse IgG (1 : 5000, Santa Cruz, USA).The membrane was then washed and immunoblotting bands were detected using the WEST-ZOL detection system (iNtRON Biotechnology, Korea).

Results and Discussion
This study was designed to demonstrate induction of HSPs in cultured RGCs by using engineered superparamagnetic Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticle agents.We employed a localized magnetic hyperthermia system by using the SPNP agents as a new nanobiotechnological approach to effectively control the local induction of HSP72.Magnetic hyperthermia using SPNP agents is considered a promising biotechnological approach to the induction of HSPs in target tissues because it can generate localized heating with accurate temperature control.Evidence suggests that manipulation of the cellular stress response may be a strategy to protect neurons from damage following cerebral ischemia [13][14][15] or during the progression of neurodegenerative conditions such as Alzheimer's disease [16,17], Parkinson's disease [18], and glaucoma [19][20][21].In this study, we demonstrate that the induction of HSPs by localized magnetic hyperthermia using   engineered SPNPs is promisingly feasible for neuroprotection modality.

Characterization of PEG 500-Coated Mn 0.5 Zn 0.5 Fe 2 O 4 .
Figure 2 shows the HR-TEM images of PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles.The PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles have a round shape with a mean size of 7.20 ± 0.85 nm and a narrow size distribution, and they are well segregated.

In Vitro Biocompatibility of PEG 500-Coated Mn 0.5 Zn 0.5 Fe 2 O 4
Particles.In vitro biocompatibility was evaluated using TEM to study the degree of particle penetration into RGCs and any cytological changes caused by this penetration.Figure 3 illustrates that the PEG 500-coated SPNPs efficiently penetrated into the cytoplasm of RGCs without causing obvious cytological changes such as cell deformation or apoptosis.In addition, we found that the SPNPs did not penetrate the nuclei of RGCs, and so did not cause nuclear fragmentation.
Our results suggest that these SPNPs can be located inside cells with no significant structural or morphological adverse effects.Furthermore, the PEG 500-coated nanoparticles inside the cells showed less agglomeration than those in our previous report [12], which may be due to the PEG coating and its weakly positively charged or uncharged surface.The results of TEM analyses of cultured RGCs suggest the potential feasibility of the coated nanoparticles as a localized hyperthermic agent for neuroprotection.magnetically induced heating.AC heating stress times varied from 600 sec to 1200 sec.The applied frequency and magnetic field were fixed at  = 110 kHz and  = 140 Oe, respectively, which are in the biologically safe and physiologically tolerable range.The PEG 500-coated nanoparticles produced stable and well-saturated heating characteristics.The temperature was controlled at a typical HSPs induction temperature of 41.0 ± 0.5 ∘ C. Figure 5 shows the AC magnetically induced heating temperature rise characteristics of RGCs treated with PEG 500coated Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles.The applied magnetic field and AC frequency were fixed at  appl = 140 Oe and  appl = 140 kHz, respectively, and the concentration of the nanofluid varied from 300 g/mL to 700 g/mL to explore the effects on HSP72 induction efficiency.As can be seen in Figure 5, the  AC,max characteristics and the corresponding AC heating response were strongly dependent on the concentration of the SPNPs (nanofluid solutions).

Induction of Heat Shock Protein by Magnetic Hyperthermia
In Vitro.In this study, 3 types of control groups were used: (1) no SPNPs and no AC magnetic field, (2) no SPNPs but AC magnetic field applied ( appl = 140 kHz and  appl = 140 Oe), and (3) RGCs incubated with SPNPs (500 g/mL) but no applied AC magnetic field.Figure 6 shows immunofluorescent staining images of HSP72 in RGCs treated with Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles after magnetic hyperthermia for 900 s at 41.0 ± 0.5 ∘ C. In this study, the RGCs were stained using FITC and DAPI with antibodies to HSP72.As can be seen in Figure 6, HSP72 was not induced in the control groups 1, 2, and 3.However, HSP72 was successfully induced in RGCs incubated with Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles and an adequate AC magnetic field.In addition, we found that the induction efficiency of HSP72 and cell death rate have a strong dependence on the temperature ramp rate, which was controlled by the externally applied AC magnetic field: cells experiencing a slower temperature elevation rate had more HSP72 induction and minimal cell loss.Specific biotechnical requirements [12] for any SPNP considered to be an agent for HSP induction are as follows: (1) it should have a high specific absorption rate, which enables it to rapidly reach the induction temperature (39-41 ∘ C) while the concentration of SPNPs is kept as low as possible; (2) it must reach the induction temperature with an applied magnetic field and frequency that fall within the ranges to ensure avoidance of bioelectrically induced damage [10] (i.e.,  < 190 Oe and  < 120 kHz); and (3) it must be highly biocompatible and should easily penetrate the target cells.Our results demonstrate that PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles exhibited heating temperature characteristics that were stable and controllable at a typical HSP induction temperature of 41.0 ± 0.5 ∘ C. Additionally, in this study, the temperature ramp rate was controlled by changing  appl within the range from 120 Oe to 160 Oe at a fixed frequency of 140 kHz.The experimental results suggest that HSPs can be successfully induced in cultured RGCs while the applied magnetic field is maintained within a biologically and physiologically safe range.

Quantitative Analysis of HSP72 Expression by Western
Blot Analysis.HSP72 expression in cultured RGCs was quantified by Western blot analysis (Figure 7).There were weak immunoreactivities for HSP72 in the control groups 1, 2, and 3.However, a strong increase of immunoreactivity was observed in RGCs incubated with Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles under the fixed AC magnetic field ( appl = 140 kHz and  appl = 140 Oe).
As can be seen in Figure 6, we confirmed that HSP72 was successfully induced in RGCs incubated with Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles when an appropriate (c) Control group 3, without magnetic field application but incubated with SPNPs (500 g/mL).(d) RGCs incubated with SPNPs (500 g/mL) and with the AC magnetic field applied ( appl = 140 kHz and  appl = 140 Oe).HSP72 was not induced in the control groups 1, 2, and 3 but was successfully induced in RGCs incubated with SPNPs when an adequate magnetic field was applied.
magnetic field was applied.The results of Western blot analyses also showed a strong increase in the immunoreactivity of similarly treated RGCs.However, induction of HSP72 was not observed in the control group 3, indicating that mechanical stress due to incubation with Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles does not contribute to the induction of HSP72.In addition, the application of only the external magnetic field without SPNP agents (control group 2) does not induce HSP72 in cultured RGCs.Furthermore, it was found that the induction efficiency of HSP72 and cell death rate are highly dependent on the temperature ramp rate.We speculate that a too rapid increase in temperature by using SPNP agents may inhibit RGCs from keeping enough thermal energy to induce stress proteins such as HSP72.
Moreover, the more rapid temperature rise might lead to higher RGC cell death during HSP induction.
Based on these results, the application of SPNP agents could be a promising biotechnical method for inducing localized HSP72.Because of specific limitations to intravenous injection of nanoparticles into the choroid, the SPNP agents can be injected into the intravitreal space of the eye.In a study performed by our group, the injected nanoparticles successfully diffused into the retina, with most found in the inner plexiform layer [12].This is the layer most adjacent to the RGCs; therefore, highly effective heat transfer to the RGCs can be expected.These findings suggest that SPNP agents could potentially be used as neuroprotective mediators in glaucoma treatment.3) Control group 3, without magnetic field application but incubated with SPNPs (500 g/mL).(4) NP(+)/MF(+) group, RGCs incubated with SPNPs (500 g/mL) and with the application of AC magnetic field ( appl = 140 kHz and  appl = 140 Oe).

Conclusions
We successfully induced HSP72 in cultured RGCs with hyperthermia via SPNPs, PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 .Considering that glaucoma ultimately causes RGC death, an endogenous neuroprotection strategy through induction of an HSP response can widen the field of glaucoma treatment.Furthermore, our results suggest that the Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles could be used as a neuroprotective treatment method for cerebral ischemia and neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease.Further studies with animal models and functional evaluations will be necessary to confirm Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles as potential novel neuroprotective therapeutic agents.

Figure 1 :
Figure 1: (a) An AC magnetic field generation system designed for both in vitro and in vivo hyperthermia studies.This system consists of AC coils, capacitors, DC power supplies, and wave generators.(b) Microcentrifuge tube containing RGCs with inserted optical thermometer.

Figure 3 :
Figure 3: Transmission electron microscopy image of RGCs treated with PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles.Nanoparticles (red circles) were found in the cytoplasm, but not in the nucleus (N), and they caused no morphological deformation.(a) Control, (b) 30 g/mL Mn 0.5 Zn 0.5 Fe 2 O 4 .

Figure 4 :
Figure 4: AC magnetically induced heating temperatures of RGCs treated with SPNPs (500 g/mL) with different holding time of AC heating from 600 sec to 1200 sec.Applied frequency and magnetic field strength were constant at 140 kHz and 140 Oe, respectively.The temperature was controlled at a typical HSPs induction temperature of 41.0 ± 0.5 ∘ C.

Figure 5 :
Figure 5: AC magnetically induced temperature rise characteristics of RGCs treated with PEG 500-coated Mn 0.5 Zn 0.5 Fe 2 O 4 nanoparticles.Measurements were taken at the fixed applied frequency of 140 kHz and magnetic field of 140 Oe, with nanoparticle concentrations ranging from 300 g/mL to 700 g/mL.This figure describes the dependence of  AC,max characteristics on SPNP concentration.

Figure 6 :
Figure 6: Immunofluorescent staining of HSP72.(a) Control group 1, without SPNPs and without magnetic field application.(b) Control group 2, without SPNPs but with magnetic field application ( appl = 140 kHz and  appl = 140 Oe).(c)Control group 3, without magnetic field application but incubated with SPNPs (500 g/mL).(d) RGCs incubated with SPNPs (500 g/mL) and with the AC magnetic field applied ( appl = 140 kHz and  appl = 140 Oe).HSP72 was not induced in the control groups 1, 2, and 3 but was successfully induced in RGCs incubated with SPNPs when an adequate magnetic field was applied.